FACTORS INFLUENCING ADVERSE EVENT AND ERROR REPORTING IN ANAESTHESIOLOGY
Steven Robert Nel
A research report submitted to the faculty of Health Sciences, University
of the Witwatersrand, Johannesburg, in partial fulfilment of the
requirements for the degree of Master of Science in Medicine in
Anaesthesiology
Johannesburg, 2017
ii
DECLARATION
I, Steven Robert Nel, declare that this research report is my own work. It
is being submitted for the degree of Master of Medicine in the University
of the Witwatersrand, Johannesburg. It has not been submitted before for
any degree or examination at this or any other university.
Signed
On this day of 2017
iii
ABSTRACT
Background
Adverse events and errors are a widespread cause of morbidity and mortality in the health
care environment. Adverse event and error reporting systems have been shown to
potentially reduce the occurrence of these events, however there is still significant under-
reporting. Little is known regarding the barriers to reporting of adverse events and errors in
the context of South Africa, or what emotional and attitudinal barriers may be present
regarding a formal reporting system amongst anaesthetists in the Department of
Anaesthesiology at the University of the Witwatersrand.
Methods
A prospective, descriptive, contextual study design utilizing an anonymous self-administered
questionnaire was distributed to 133 anaesthetists who attended academic anaesthetic
meetings.
Results
One hundred and eighteen questionnaires met the criteria for analysis, giving a response
rate of 92%. Barriers to reporting included a “code of silence” in medicine and blame from
colleagues. If a specified error as opposed to an adverse event had occurred, participants
were more likely to agree with barriers regarding fear of litigation, disciplinary action, getting
into trouble, as well as colleagues that may be unsupportive. Strategies to promote reporting
of adverse events and errors include senior role models who encourage reporting and
individualised feedback regarding reports made.
iv
Conclusions
Most anaesthetists in our study disagreed with barriers to reporting an unspecified adverse
event. However, if an error has occurred, reporting behaviour may be inhibited by barriers
regarding fears of litigation, disciplinary action and lack of support. Senior role models that
openly support reporting along with individualised feedback may increase reporting rates.
v
ACKNOWLEDGEMENTS
My thanks go to the following people:
� my supervisors, Juan Scribante, Helen Perrie, and Professor Lundgren for all their help and support
� my wife and family for their help and support.
vi
TABLE OF CONTENTS
DECLARATION .......................................................................................................... ii
ABSTRACT ................................................................................................................ iii
ACKNOWLEDGEMENTS ........................................................................................... v
TABLE OF CONTENTS ........................................................................................... vi
LIST OF FIGURES ..................................................................................................... ix
LIST OF TABLES ....................................................................................................... x
LIST OF ABBREVIATIONS ....................................................................................... xi
SECTION 1: Literature Review ................................................................................. 1
1.1 The safety of anaesthesia as a specialty .............................................................. 1
1.2 The historical context for reporting of adverse events and errors .................... 2
1.3 Incidence of adverse events and errors in clinical practice ............................... 5
1.4 Current reporting structures .................................................................................. 7
1.5 Benefits associated with reporting of adverse events and errors ..................... 8
1.6 Limitations associated with reporting of adverse events and errors .............. 11
1.7 Factors influencing adverse event and error reporting .................................... 13
1.8 References ............................................................................................................. 21
vii
SECTION 2 – Journal guidelines to authors ......................................................... 25
SECTION 3 – Draft Article ....................................................................................... 31
Abstract ............................................................................................................................ 32
Introduction ...................................................................................................................... 33
Method .............................................................................................................................. 34
Results .............................................................................................................................. 36
Discussion ........................................................................................................................ 42
Limitations ........................................................................................................................ 44
Conclusion ....................................................................................................................... 44
Acknowledgements ......................................................................................................... 45
Conflict of Interest ........................................................................................................... 45
References ....................................................................................................................... 46
SECTION 4 - Appendices ........................................................................................ 49
SECTION 5 - Proposal ............................................................................................. 52
5.1 Introduction ........................................................................................................... 53
5.2 Problem statement ................................................................................................ 54
5.3 Aim ......................................................................................................................... 54
5.4 Objectives .............................................................................................................. 55
viii
5.5 Research assumptions ......................................................................................... 55
5.6 Demarcation of study field ................................................................................... 56
5.7 Ethical considerations .......................................................................................... 56
5.8 Methodology .......................................................................................................... 57
5.9 Collection of data .................................................................................................. 58
5.10 Significance of the study ...................................................................................... 60
5.11 Validity and reliability of the study ...................................................................... 60
5.12 Potential limitations of the study ......................................................................... 61
5.13 Project outline ....................................................................................................... 62
5.14 Financial plan ........................................................................................................ 62
5.15 References ............................................................................................................. 63
ix
LIST OF FIGURES
Figure 1: Attitudinal and emotional factors regarding the reporting of adverse events 38
Figure 2: Strategies that may promote or improve the reporting of adverse events and errors 41
x
LIST OF TABLES
Table 1: Factors that impede physician disclosure of medical errors (53) 17
Table 2: Factors that facilitate physician disclosure of medical errors (53) 18
Table I: Summary of questionnaire structure 35
Table II: Participant’s demographic data 37
Table III: Current reporting structures available 37
Table IV: Potential barriers to reporting an adverse event or error when anaphylaxis occurred
in a given clinical scenario 40
xi
LIST OF ABBREVIATIONS
ANZCA – Australian and New Zealand College of Anaesthetists
MeSH – Medical Subject Headings
NASA – National Aeronautics and Space Administration
NRLS – National Reporting and Learning System
SASA – South African Society of Anaesthesiologists
UK – United Kingdom
URL – Uniform Resource Locator
US – United States (of America)
WHO – World Health Organisation
WITS – University of the Witwatersrand
1
SECTION 1: Literature Review
This section aims to review the current literature regarding adverse event and error reporting
in general medical practice, as well as in the speciality of anaesthesia. Initially, the safety of
anaesthesia as a speciality will be explored. The review will then analyse the historical
context in which adverse events and errors were reported, whilst reviewing problems with
definitions and taxonomy of adverse event and error reporting systems. The review will then
contrast the incidence of adverse events and errors in general medical practice as compared
to anaesthesia. Current reporting structures will be reviewed, followed by analysis of the
literature with regards to limitations of these reporting structures. Lastly, the factors that may
influence the reporting of adverse events and errors will be reviewed.
1.1 Thesafetyofanaesthesiaasaspecialty
“Health care in the United States is not as safe as it should be – and can be. At least 44,000
people, and perhaps as many as 98,000 people, die in hospitals each year as a result of
medical errors that could have been prevented…” states an excerpt from the report “To err is
human: Building a safer health system” released in 1999 by the Institute of Medicine. (1)
The report highlighted how there was an “epidemic of medical errors” occurring in health
care, and advocated for the delivery of safer care for patients. This led to a strong political
response from the United States (US) Congress who allotted $50 million to the agency for
Healthcare Research and Quality with the main focus of reducing medical errors. (1)
Even though there is renewed emphasis on safe health care as a result of the report from
the Institute of Medicine, anaesthesia decades previously had implemented measures to
reduce harm to patients in the operating room. Anaesthesia is well known for advancements
in patient safety, and is described as one of the leaders with regards to reducing mortality,
having some of the lowest death rates. (2) However, this was not always the case.
The landmark study by Beecher and Todd in 1954 (3) described the rate of anaesthetic
related deaths and complications in nearly 600 000 anaesthetics given over a five year
period, and was one of the first recorded scientific studies aimed at reducing the risks
associated with anaesthesia. The study found primary anaesthetic related mortality rates of
1 in 1560 patients. This resulted in various changes in practice, most notably the suggestion
for the removal of curare, a muscle relaxant that was associated with a higher mortality. This
2
study replaced previous anecdotal evidence that was the basis for clinical decision-making
in anaesthesia (3).
Haller (4) cites evidence from the literature detailing the decline in the anaesthesia related
mortality rate. He states that at the end of the 19th century 1 in 900 patients died subsequent
to anaesthesia. He contrasted this with a substantial reduction in the current mortality rate of
between 1 in 100 000 to 1 in 200 000. He does however caution that the above statistics
should be interpreted with care, as there is still not currently a clearly defined taxonomy of
“anaesthesia related death”, and thus the figures could potentially be different to what is
described.
Gaba (2), an anaesthesiologist and a pioneer in human factors related to anaesthesia,
explored reasons for improved safety in anaesthesia. With anaesthesia becoming a more
complex specialty, and the fact that it extended to include intensive care, he proposed that it
attracted a “higher calibre of staff”. He also stated that due to the fact that anaesthesia has
no actual therapeutic benefits for patients and is a relatively high-risk specialty, that
anaesthetists tend to be reluctant to take risks and are more inclined to focus on safety of
patients. He further proposed that as the field of anaesthesia attracted individuals with a
background in biomedical engineering, safety models were imported from other “hazardous
technological pursuits, including aviation”. Lastly, he observed that in the 1970’s to 1980’s
the cost of medical malpractice insurance climbed dramatically, and “was at risk of becoming
unavailable”, possibly promoting a shift towards safer anaesthetic practices. (2)
As safe as anaesthesia is purported to have become, there is still a risk of morbidity and
occasionally mortality (4-6). Botney et al (7) described some of the potentially avoidable
complications that could occur in anaesthetic practice, and detailed the risk of occurrence.
Morbidities like dental injury (1 in 4500) and intra-operative awareness (1 in 500) were still
reasonably common occurrences, whereas morbidities like neurological injury and airway
injuries were rare (1 in 5000-10 000).
1.2 Thehistoricalcontextforreportingofadverseeventsanderrors
The various methods by which patient safety has been advanced in anaesthesia were
reviewed by Runciman and Merry (8) in their book titled “The Wondrous Story of
Anesthesia”.
3
The authors highlighted how:
• inthe1960’sRossHollandinAustraliaandGaiHarrisoninSouthAfricaresearched
mortalityduetoanaesthesia,showingatrendtowardsmortalityreductionoverthe
nextfewdecades.
• in1974Cooperinitiatedthecriticalincidentreportingsysteminanaesthesia,and
alsodescribedhis1978essayonpreventableanaestheticmishaps.
• inthe1980’sthemedicalindemnitycrisisresultedinthecreationoftheAnesthetic
PatientSafetyFoundationin1985,whichformedasaresultofthe1984
InternationalCommitteeforthePreventionofAnaesthesiaMortalityandMorbidity
meetinginBoston.
• in1988theNationalConfidentialEnquiryintoPerioperativedeathsintheUnited
Kingdom(UK)andAustralianPatientSafetyFoundationformed.
• in1991theretrospectivemedicalrecordreviewsintheHarvardMedicalPractice
studyledtothedevelopmentofa“comprehensiveclassificationofthingsthatcan
gowrong”,creatingthefoundationforthemodernincidentreportingsystemsof
today.
The preliminary work of Flanagan (9), in 1954, was where the term “The critical incident
technique” was first introduced into the literature. It arose from his observations of the
Aviation Psychology Program of the United States Army Air Forces in World War II. He
described how the aviation industry used studies to analyse various mishaps in fighter pilots,
and fully discussed the method by which a critical incident may be detailed. He further
described a critical incident as “any observable human activity that is sufficiently complete in
itself to permit inferences and predictions to be made about the person performing the act”.
Cooper et al (10) in 1978 and Williamson et al (11) in 1985 were the first discernible role
players in the anaesthetic environment to review undesirable anaesthetic outcomes as
“critical incidents”, and were the first to consider a framework from which other anaesthetists
may learn from these incidents. The first notable critical incident monitoring system was set
up in 1988 in Australia when Runciman (8) spearheaded the movement to create the
Australian Incident Monitoring Study in Anaesthesia.
4
This system was the catalyst for the formation of the Australian Patient Safety Foundation in
1989, which took a broader focus on patient safety in Australia (12). It was seven years later
that the foundation created the Australian Incident Monitoring System as an “initiative to look
at options for reducing risk in South Australian health care units”. (12)
Since then, there has been international progress in the formation of other national incident
reporting systems, examples being the National Reporting and Learning System (NRLS) in
the UK, created in 2003 (13), and the Critical Incident Reporting System that has been
utilized in Switzerland since 1997 (14).
A discussion of other incident reporting systems will follow later in the chapter.
Definitions of adverse events and errors
There is currently no known universal taxonomy for anaesthetic terms in incident reporting
systems (15), and definitions of adverse events and errors are numerous and variable (4, 5,
15-18). Staender (17) argues that defining “critical incidents” or “adverse effects” can be
difficult due to the variability of outcomes from each incident. Smith et al (5) state that
without agreement on definitions, incident reporting systems in healthcare are unlikely to
reach the same potential as in other industries.
The World Health Organization (WHO) draft guidelines for adverse events and reporting
systems (19) define adverse events as: “an injury related to medical management, in
contrast to complications of disease. Medical management includes all aspects of care,
including diagnosis and treatment, failure to diagnose or treat, and the systems and
equipment used to deliver care. Adverse events may be preventable or non-preventable”.
The WHO draft guidelines (19) further describe an error as: “the failure of a planned action
to be completed as intended (i.e. error of execution) or the use of a wrong plan to achieve an
aim (i.e. error of planning). Errors may be errors of commission or omission, and usually
reflect deficiencies in the systems of care”.
Smith et al (5) state that the UK Royal College of Anaesthetists official definition of a critical
incident was well known in their study population, yet staff interviewed created their own
working definitions based on perceived seriousness of the event, as well as eventual
outcome, with good outcomes less likely to be reported. They suggest that the experience of
the anaesthetist determines their “definitional power”, where a more experienced
anaesthetist may judge an adverse event as routine practice, whereas a less experienced
5
anaesthetist may judge the same adverse event as an incident worth reporting. The example
given being the disconnection of a circuit while a patient was under general anaesthesia.
Tamuz et al (16) focused on the impact definitions may have on error reporting rates in their
2004 study titled “Defining and classifying medical error: lessons for patient safety reporting
systems”. The authors showed how without clear definitions of errors or adverse events
being available, incidents were “defined away” and not reported, as they did not meet the
“working definition of an error”.
1.3 Incidenceofadverseeventsanderrorsinclinicalpractice
General incidence of adverse events and errors
Landmark research by Brennan et al (20) had appraised the medical records of 30 121
patients admitted to 51 acute care hospitals in New York in 1984. They reported an adverse
event rate of 3.7% of all admissions. There was further research done on the same data
used by Brennan et al (20), which found that the nearly 70% of the reported adverse events
were due to medical error (21).
In 1995, Wilson et al (22) reviewed medical records of 14 179 patients admitted to 28
hospitals, in order to understand the quality of health care in Australia. They found that the
rate of adverse events was 16.6 per 100 admissions, with “permanent disability in 13.7%”
and “death in 4.9%”. They also determined that at least 50% of the adverse events were
preventable.
Using computerised models to detect adverse drug events at a hospital in Utah, United
States in 1991, Classen et al (23) found rates of adverse drug events to be 1.7%. This
contrasts with Bates et al (24) in 1995 who found substantially higher rates of adverse drug
events, totalling 6.5% of all admissions. This was achieved by using chart reviews and self-
reports from health care professionals.
Andrews et al (25) used ethnographers to observe daily activities in 10 surgical units
throughout three academic hospitals over a period of two months in the US. They found that
adverse event rates were 45.8% (480 of 1047 patients), of which 17.7% (185 patients) of
adverse events were of a serious nature.
Starmer et al (26) performed a prospective intervention study at a Boston Children’s Hospital
in 2013, where they implemented a resident handoff bundle with standardised
communication and handoff training. They had 1255 admissions over the period of the
6
study, of which 642 were in the pre-intervention group and 613 in the post-intervention
group.
They found initially that there were error rates of 33.8% in the pre-intervention group of
patients, with preventable adverse event rates of 3.3%. Their post–intervention results
showed a 50% decline in both rates. (26)
Anaesthetic related incidence of adverse events and errors
Cooper et al (10) studied preventable anaesthesia mishaps in 1978 in the US, and found
that 18% of anaesthetics would develop “an unexpected problem requiring intervention”
during the anaesthesia, and “3 to 5% of those anaesthetics will involve a serious unplanned
event”.
Catchpole et al (27) analysed 12 606 anaesthetic related incidents over a two year period in
UK based hospitals, and found that 75% of the incidents resulted in no harm, 22.5% in little
or moderate harm, and 2.1% in severe harm or death.
These results contrast with a 2009 study by Gupta et al (28) in India, which was a
prospective internal audit into critical incident reporting in anaesthesia. The study was
performed over a one-year period with a total of 14 134 anaesthetics being delivered to
patients. The results of the study show that 112 (0.79%) critical incidents were reported, with
complete recovery in 80 (72%) patients and death in 32 (28%) patients. They also found that
incidents were highest amongst paediatric, American Society of Anesthetists physical class
grade 1, general surgery patients undergoing emergency surgery.
Gupta et al (28) further showed that the incidence of mortality was found to be 22.6 per 10
000 anaesthetics. With anaesthesia as the only cause of death, the rate was 5.6 per 10 000
anaesthetics. The main reason stated for the death rate was human error (75%), which was
predominantly ascribed to poor judgment (67.5%).
Lundgren (29) in 2011 analysed the perioperative death rates in two major academic
hospitals in Johannesburg. The author found that anaesthetic contributory death rate was
approximately 0.4 per 10 000 anaesthetics delivered, which was described as being
comparable to the rates in the UK. This contrasts with Gupta et al (28), where their
described mortality rate in a tertiary hospital was more than ten times that shown by
Lundgren (29).
7
Madzimbamuto and Chiware (30) described the implementation of a critical incident
reporting system in anaesthesia in two teaching hospitals in Zimbabwe. They had 62
completed critical incident forms submitted between May and October in 2000, with a total of
14 165 anaesthetics delivered. They described 130 critical incidents, giving a rate of 0.92. Of
these 40 were emergency cases and 22 were elective cases. Incidents reported were
“hypotension, hypoxia, bradycardia, electrocardiogram changes, aspiration, laryngospasm,
high spinal and cardiac arrest. They reported that human error was responsible for 50% of
critical incidents, and equipment failure a further 50%. Patient outcomes showed that 15%
died, and 23% required unplanned admission to an intensive care unit setting. The mortality
rate was comparable to that described by Gupta et al (28).
Gordon et al (6) published a survey in 2006 by South African anaesthetists with regards to
drug administration errors. They sent confidential surveys to 720 anaesthetists nationally,
and had 133 returned surveys. The authors found that 94% of anaesthetists in the sample
had inadvertently administered an incorrect drug to a patient, with five deaths and three non-
fatal cardiac arrests noted following incorrect drug administration.
Labuschagne et al (31) performed a similar survey in 2011 in the Free State. They found
that 39.3% (n=84) of participants had at least one drug administration error in their careers.
1.4 Currentreportingstructures
Various systems for reporting of adverse events and errors have been employed within the
health care industry, each having an impact on a professional’s likelihood of reporting an
adverse event or error (32).
Pham et al (33) in 2013 published a review article identifying different structures and
functions of incident reporting systems. The following examples were given:
• nationalreportingsystems(AustraliaIncidentMonitoringSystemandtheNRLSin
theUK)
• localsystems(PatientSafetyNetworkandthePennsylvaniaSafetyReporting
System)
• specialityspecific,anexamplegivenbeinganIntensiveCareUnitReportingSystem
• incidentspecific,suchastheMedMARxsystemwhichfocusesonmedication
incidents.
8
Suresh et al (34) in 2004 suggested that amassing errors that are speciality specific might
be more likely to show a trend of errors, than if the errors were diluted in a general error
reporting system, with the consequent possibility of being overlooked.
Leape (35) argues that many of the current error reporting systems are inadequate, as they
are not meeting the main objective of improving patient safety. The author analysed the
success of the NASA Aviation Safety Reporting System as portrayed by Billings (36), and
suggested three factors why medical incident reporting systems were inadequate. Firstly,
reporting needed to be safe, for example, pilots are “immune from disciplinary action if they
report promptly”. The second factor was that they needed to be simple, indicating that a one-
page report was ideal. Lastly, that incident reporting systems needed to be made
worthwhile, citing an example of how in aviation, experts scrutinised the confidential report
and then distributed recommendations to pilots and the Federal Aviation Administration. (35)
Leape (35) further gave examples of successful voluntary reporting medical error systems,
namely the Medication Error Reporting Program, MedMARx and the National Nosocomial
Infection Survey, describing how these systems shared similar features with the NASA
Aviation Safety Reporting System.
Leape (35) contrasted the success of voluntary reporting systems to those that are
mandatory, and stated that mandatory systems are “seldom simple, safe, or worthwhile”.
This is in opposition to the views of Kohn et al (1) whose report suggested that mandatory
reporting was an important part of patient safety culture, as it ensured that health care
practitioners and institutions were held responsible for their actions.
Holden et al (32) reviewed anonymity and confidentiality in designing an incident reporting
system. They described factors that might allow for a more effective reporting system, with a
suggestion that a system that embraces anonymity would allow for better reporting of
incidents, as they will be non-punitive. Runciman (37) concurs than anonymous reporting
would be beneficial as health care staff would be more likely to report adverse events or
errors, even though he conceded that ethically it was contentious.
1.5 Benefitsassociatedwithreportingofadverseeventsanderrors
“The belief that one day it may be possible for the bad experience suffered by a patient in
one part of the world to be a source of transmitted learning that benefits future patients in
many countries is a powerful element of the vision behind the WHO World Alliance for
Patient Safety.” – Sir Liam Donaldson, chair for the World Alliance for Patient Safety. (19)
9
There are many reports and studies on the benefits associated with incident reporting
systems available in the literature (17, 33, 38-45), many of which echo the hopes described
by Donaldson.
Smith and Forster (38) proposed that there would be “improved professional learning and
better patient outcomes from higher quality care”. They also described the potential for
improved trust between staff and patients, fewer chances of litigation, and a “more realistic
view by patients of staff and medicine’s limitations”
Wood and Nash (40) analysed mandatory state-based error reporting systems in the United
States in 2005, and found that a collection of data on adverse incidents would allow for
analysis regarding trends and similarities. This allowed for identification of common
incidents, and focused attention for system improvements to occur.
Williams and Osborn (41) in 2006 suggested the opportunity to share the analysis of
adverse events globally, thus allowing for cross border learning, and potentially reducing
adverse event occurrence.
According to Staender (17) there are meaningful lessons that can be learnt from incident
reporting systems. He described four advantages to incident reporting. These were quantity,
recoveries, root cause information, and learning.
With regards to quantity he argued that incidents are much more common than severe
events, and thus there will be many more incidents to analyse for any given period of time.
He also described how recoveries could be analysed by looking for preventative processes
whereby the serious events were avoided, rather than analysing what had actually
happened in the case of a severe event. He further described how incidents could be
analysed from their root causes, thus forming “the basis of very strong quality-improvement
actions”. Lastly, he theorized that by passing on personal experiences on critical events,
others might learn valuable lessons, thus possibly avoiding similar events.
Smith et al (5) echoed Staender’s (17) proposed advantage of learning through reporting.
Their study was of a qualitative design, and explored adverse events in anaesthetic practice,
based on focus group interviews with medical professionals. They depicted a scenario in a
focus group, where they asked practitioners how they had gained their anaesthetic
knowledge. They reported that this prompted many anaesthetists to focus on cases that
“had not gone according to plan”. They described how this “triggered a lively discussion” of
10
learning from events that had occurred in an anaesthetist’s career, and concluded that
reporting was beneficial for continuing education within the anaesthetic community.
Staender’s (17) opinions also support those of Vincent (45), who described how the analysis
of incidents could provide a framework on which to base decisions regarding policy and
practice changes that could ultimately lead to reduced occurrences of adverse events.
A review by Pham et al (33) further elaborated on Staender’s (17) idea that global learning
can occur from incident reporting systems. They described how incident-reporting systems
could benefit multiple organisations, through sharing of information at “local, regional,
national and international levels”. They use the example of the Canadian Global Patient
Safety Alerts, which is a collection of adverse events with full case details. These safety
alerts describe the processes involved through which the failure occurred, and discuss the
interventions implemented to prevent a further occurrence of the same error. The Chief
Executive Officer of the Canadian Patient Safety Institute cites these safety alerts have
“more than 684 alerts and 3400 recommendations from 23 contributing organizations around
the world” (42). This information has been used worldwide to alter safety practices in
medicine.
Pham et al (33) also observed that incident reporting systems tend to alter the patient safety
culture of the organization, especially if the incidents were reported back to the health care
professionals, as it communicated how important the organization felt patient safety was.
Kaplan and Barach (44) considered the evaluation of “near misses” rather than adverse
events. In doing so they proposed that analysis of near misses had many benefits such as:
• “nearmissesoccur300-400timesmorefrequently[thanadverseevents],enabling
quantitativeanalysis
• therearefewerbarrierstodatacollection,allowinganalysisofinterrelationshipsof
smallfailures
• recoverystrategiescanbestudiedtoenhanceproactiveinterventions
• hindsightbiasismoreeffectivelyreduced”.
As echoed by Pham et al (33), Kaplan and Barach (44) also stated “perhaps the least
appreciated and most unique attribute is the potential for incident reporting to engage the
staff in safety activities. This involvement may bring about mindfulness and a change in
safety culture”.
11
Furthermore, there are also benefits to society with incident reporting systems. In 2000
Barach and Small (43) described societal incentives to reporting. They argued that from a
legal perspective, reporting would ensure accountability and enforce reporting statutes
amongst health care professionals. From a regulatory perspective they claimed there could
be enhanced regulatory trust, greater transparency and more public accountability.
1.6 Limitationsassociatedwithreportingofadverseeventsanderrors
As beneficial as incident reporting systems could potentially be, they are also associated
with several limitations.
Catchpole et al (27) analysed 12 606 reported anaesthesia incidents from the UK NRLS.
They endeavoured to understand how a large database like the NRLS might contribute to
identifying incidents and system based problems. As part of their analysis they discussed
some of the limitations such a database may have. The authors concluded that the “lack of
detail inherent to large data fields prohibit translation of these results into robust arguments
for immediate change in clinical practice”. (27)
Catchpole et al (27) also found that medication error or treatment failure may be difficult to
tell apart, especially in anaesthesia with the amount of medications given to any one patient.
They referred to one example where an extravasation injury was classified as a medication
error, showing problems with hierarchy in the taxonomy of medical error reporting. (27)
Henriksen and Kaplan (46) discussed issues regarding hindsight bias in their 2003 review.
They described hindsight bias as “the exaggerated extent to which individuals indicate they
would have predicted the event beforehand”. They gave evidence regarding hindsight bias
and its effects on medicine and health services research. They discussed that the
retrospective analysis of medical errors, as is done with incident reporting systems, is
particularly prone to hindsight bias, and describe how this might have a negative influence
on the analysis of the error.
Another limitation to incident reporting systems is the high prevalence of under-reporting
amongst health care professionals. According to Taylor et al (47) in 2004, only one in five
incidents are reported. They also showed that nurses are 80% more likely to report errors
than physicians (OR = 2.8, 95% CI: 1.3 – 6.0). Under-reporting of incidents can make the
overall analysis less meaningful (27).
A further limitation can be found in the study by Braithwaite et al (48), in Australia, in 2008.
They analysed attitudes towards the implementation of an incident reporting system. The
12
survey took the form of an online anonymous questionnaire completed by 2185 health
practitioners, which found that nurses tended to report more adverse events than physicians.
The authors argued that this disproportional reporting might lead to bias and the potential for
prioritisation of certain incidents over others.
Catchpole et al (27) also described the potential for data inaccuracies when incidents were
reported by healthcare professionals who were not directly involved in the events that
occurred. An example given was the reporting of epidural related complications to the
NRLS, where it was found that the higher levels of harm reported for epidurals might “reflect
a bias away from reporting no-harm events”. These errors may have been reported by
nurses, who may not have understood the hierarchical classification of errors related to
epidurals. Thus, many incidents were reported as “other” and as “treatment/procedure”
errors. These errors may have been correctly classified had the physicians directly involved
reported them.
Catchpole et al (27) concluded that for data from an incident reporting system to be
meaningful, there needs to be a “sufficiently validated reporting and analysis framework”,
without which “substantial classification inconsistencies” must be assumed.
Tamuz et al (16) reported similar concerns in their study on defining and classifying medical
error, where they analysed how medication errors were reported in a hospital complex.
They found an important source of underreporting was due to potential medication incidents
being “defined away” or “classification bias”. They described “defined away” as the
respondent’s definition of what the error was not meeting the didactic definition of the
pharmacy’s error reporting system. This confusion around definitions thus resulted in the
staff being less likely to report it, as the incident did not classify as an “error”.
Woolf et al (49) studied the effect “cascade analysis” had on analysis and prevention of
medical errors. They found that there was difficulty with assigning a single classification
when a “cascade” of errors had occurred. This also led to similar classification limitations as
discussed by Tamuz et al (16).
These and other problems with classification were summarised in a review of medical error
reporting systems by Holden and Karsh (32) in 2007. They emphasised the importance of
developing “mature taxonomies in health care”. This would allow for conforming standards
on what to report, the format the report should take, and, ultimately will facilitate “meaningful
analysis and control steps in the safety process”.
13
Pham et al (33), in 2013, describe various limitations to incident reporting systems. Firstly,
they state that incident reporting systems are non-random samples encompassing all
potential reports, and thus cannot be used to measure patient safety rates. Reasons stated
in their review included the fact that events or errors were under-reported, citing reporting
rates as low as 7% of all incidents, as well as the fact that different methods are used to
detect errors resulting in varying conclusions about safety. The authors also stated that
incident reporting systems could not be used to compare organisations, as institutions that
report more errors did not necessarily have higher error rates.
According to Pham et al (33) incident reporting systems could not be used to measure
changes over time with regards to safety practices. The reason for this is that “valid error
rates are required to make inferences in safety over time”, which cannot be ascertained with
incident reporting systems. The authors suggested that incident reporting systems can
generate too many reports, citing an example of how Johns Hopkins Hospital generates
approximately 500 reports per month. They state that it may not be possible for an
organisation to analyse such a high number of reports, and even less likely to create
actionable responses. (33)
Lastly, Pham et al (33) indicate that “incident reporting systems often do not generate in-
depth analysis or result in strong interventions to reduce risk”. Reasons described were that
with limited resources, error investigations were often superficial, and that staff have “limited
or no training in adverse event investigation or human factors”. Furthermore, meaningful
system changes are rare, with most interventions being “informing staff involved and
education/training”. (33)
1.7 Factorsinfluencingadverseeventanderrorreporting
Factors which may impede reporting
As discussed previously, the rate of under-reporting of adverse events and errors is
inadequate (16, 33, 47). Some of the reasons for this will be explored.
Holden and Karsh (32) compiled a review of medical error reporting system design
considerations in 2007, and described various barriers to reporting.
They discussed how busyness and fatigue play an important role, as health care
practitioners could have high work burdens. Suresh et al (34) described the leading self-
reported barriers to reporting to be “time involved in documenting an error” and “extra work
involved in reporting”.
14
Holden and Karsh (32) also described how lack of knowledge about the reporting system
acted as a barrier to reporting. They stated that medical staff might be unaware that
reporting structures exist in their institutions, or not know their purpose. Jeffe et al (50) used
focus groups to explore physicians’ and nurses’ perspectives on error reporting in hospitals.
They found that there was some confusion over whether or not less serious errors and near
misses needed reporting, and whose responsibility it was to report these.
Heard et al (51), in 2012, surveyed 443 anaesthesiologists and anaesthesiology residents in
Australia and New Zealand with an anonymous questionnaire. The authors analysed the
effect of the “perfectibility model” on the likelihood of reporting. They described the
“perfectibility model” as being “based on beliefs that physicians are capable of and should
provide error-free practice, with anything less being unacceptable”. They hypothesised that
this model would have a negative impact on the anaesthesiologist’s likelihood to report
medical errors. However, they found that 79% of respondents disagreed or strongly
disagreed with the statement: “If a doctor is careful enough, he or she will not make an
error”. This contrasts with a systematic review by Kaldjian (52) who showed that medical
perfectionism is a reason for reduced reporting of errors.
Even though 79% of respondents disagreed with the statement in Heard et al’s (51) study
regarding the perfectibility model, nearly 30% agreed or strongly agreed that: “If I admit to an
error I will feel like a failure”, and “It would affect my self-esteem to admit to an error”. These
statements contextualise what many other authors have found regarding reasons for not
reporting (32, 43, 52, 53). They have found that blame and shame are serious disincentives
to reporting adverse events and errors. This may explain why in the study by Heard et al
(51) 5% of respondents agreed or strongly agreed that they would “protect their self-interests
ahead of a patient’s; for example, by hiding or denying an error”. Additionally, 10% of
respondents agreed or strongly agreed that “they would cover up an error if they could”.
The authors stated that the results may be influenced by “social desirability bias”, and thus
respondents answered the questions “as a good person should” rather than answering what
they would actually do in a scenario with a bad outcome. To account for this, Heard et al
(51) included a second section to their questionnaire where they analysed how respondents
would react to an adverse event with or without an error. The respondents were given a
scenario where anaphylaxis occurred due to the giving of an antibiotic intra-operatively. In
half the respondents the questionnaire indicated that the anaphylaxis occurred even though
the patient had no history of allergy to any antibiotics, and in the other half the questionnaire
indicated that the doctor knew that the patient had an allergy to the specific antibiotic, and
15
inadvertently had given it. The authors found that anaesthesiologists in the “error” group
were more likely to agree with barriers regarding “litigation, disciplinary action, trouble,
blame, lack of support from colleagues, and not wanting the case discussed in meetings”,
than those in the non-error group. (51)
Leape (35) also supported the finding that doctors and nurses are fearful of litigation and
disciplinary action, and are thus less likely to report adverse events and errors. The author
also reported that among physicians “shame and fear of liability, loss of reputation, and peer
disapproval” were especially prevalent reasons why adverse events and errors were not
reported.
Kaldjian et al (53), in 2006, developed an “empirically derived, comprehensive taxonomy of
factors that affect voluntary disclosure of errors by physicians”. They reviewed 316 articles,
and identified 91 impeding or facilitating factors to the willingness of physicians to disclose
errors. They also conducted focus group studies and found a further 27 factors that were
pertinent. These factors were collated together into domains, four domains being facilitating
factors and four domains being factors that impeded reporting of errors. (53) The four
domains that impede reporting are shown in Table 1, which was transcribed directly from
Kaldjian et al (53).
The main theme present in the first domain of “attitudinal barriers” was that of perfectionism
and the competitive nature of doctors, which was a potential cause of reduced reporting. A
persistent theme that was present in the second domain of “helplessness” was that
respondents wanted to know what was going to happen to their report, as they were fearful
that it would jeopardise them at a later stage in their careers with detrimental effects. The
third domain addressed feelings of uncertainty, and the fourth domain focused on “fears and
anxieties”. This last domain explored feelings of shame and guilt, as well as fears of
disclosing the error. (53) This last domain echoes similar findings by Wu. (54) The author
was of the opinion that medical errors have a second victim, namely the health care
professional involved, and commented that some physicians “are deeply wounded, loose
their nerve, burn out, or seek solace in drugs and alcohol”. Similar findings were also
described by Leape (35) and Heard et al (51).
Heard et al (51) uncovered similar themes. They found that the statement “doctors who
make errors are blamed by their colleagues” to be significant, where 46% of respondent
agreed/strongly agreed with the statement. They also stated that after reporting a serious
error, physicians reported high levels of “emotional distress, shame, guilt, self-reproach, self-
perceptions as failures, fear of blame and criticism…”
16
Legal concerns of reporting and financial liabilities were also profound barriers to reporting,
as shown in the fourth domain (53), a concern supported by Leape (35).
Kaldjian et al (53) conclude that by taking into account factors that could impede reporting
adverse events and errors, implementation of reporting structures may be more likely to
succeed.
17
Table 1: Factors that impede physician disclosure of medical errors (53)
Attitudinal Barriers Helplessness
• Perpetuating perfectionism, and blaming
and humiliating those involved with errors
• Perpetuating silence about errors,
denying errors, or believing others don’t
need to know about one’s errors
• Being arrogant and proud
• Placing self-interests before patient-
interests
• Allowing competition with peers to inhibit
disclosure
• Believing disclosure is an optional act of
heroism
• Doubting the benefits of disclosure
• Lacking control of what happened to
information once it is disclosed
• Lacking confidentiality or immunity after
disclosure
• Lacking institutional and collegial support
after disclosure or a professional forum
for discussion
• Believing error reporting systems
penalise those who are honest
• Lacking feedback after reporting errors
• Lacking time to disclose errors
• Feeling helpless about errors because
one cannot control enough of the system
of care
Uncertainties Fears and Anxieties
• Being uncertain about how to disclose
• Being uncertain about which errors
should be disclosed
• Being uncertain about the cause of an
adverse event
• Disagreeing with a supervisor or trainee
about whether an error occurred
• Fearing legal or financial liability
• Fearing professional discipline, loss of
reputation, loss of position, or loss of
advancement
• Fearing patient’s or family’s anger,
anxiety, loss of confidence, or termination
of physician-patient relationship
• Fearing the need to admit actual
negligence
• Fearing the need to disclose an error that
cannot be corrected
• Fearing the possibility of looking foolish
in front of junior colleagues and trainees
• Fearing negative publicity
• Fearing the possibility of ‘fallout’ on
colleagues
• Feeling a sense of personal failure and
loss of self esteem
18
Factors which may facilitate reporting
There is less evidence in the literature regarding factors that may facilitate the reporting of
adverse events and errors than there is regarding the barriers to reporting (55). The four
domains from Kaldjian et al (53) that facilitate reporting of errors are shown in Table 2, which
was transcribed directly from Kaldjian et al (53).
Table 2: Factors that facilitate physician disclosure of medical errors (53)
Responsibility to Patient Responsibility to Profession
• Desire to communicate honestly with
patients or explain the circumstances
of an error
• Desire to show respect for patients or
treat patients fairly
• Desire to facilitate further medical care
for harmed patients
• Desire to share lessons from learned
errors
• Desire to serve as a role model in
disclosing errors or breaking bad news
• Desire to strengthen inter-professional
relationships and build inter-
professional trust
• Desire to change professional culture
by accepting medicine’s imperfections
and lessen the focus on managing
malpractice risks
Responsibility to Self Responsibility to Community
• Desire to account for one’s actions
• Sense of duty as a physician
• Desire to maintain one’s integrity
• Desire to treat others as one would like
to be treated
• Desire to empathise and apologise
• Desire to alleviate guilt or pursue
forgiveness
• Willingness to accept one’s fallibility
and limitations, and to be vulnerable
• Desire to follow one’s conscience or
‘do the right thing’
• Desire to follow one’s religious/spiritual
beliefs
• Desire to enhance the health of future
patients
• Desire to sustain patients’ trust in the
medical profession
• Desire to foster physician-patient
relationships that can absorb the shock
of error
• Desire to help patients to be more
realistic about medicine’s imperfections
• Desire to help patients understand the
complex causes of errors
The “responsibility to patient” domain centres on the respect that a physician should have for
a patient as a fellow human, and suggests that free and honest communication is central to
19
continuing a respectful physician-patient relationship. The second domain focuses on the
responsibility the physician has towards him/herself, and explores “professional and
personal values” derived from the physician’s character principles. The main theme present
was that the physician needs to have the courage and conviction to admit to making an
error, and be prepared for the associated consequences. In the domain of “responsibility to
profession”, the physician may feel that a duty is present to allow others to learn from their
error, potentially encouraging and environment where errors can be disclosed without fear of
reprisal from colleagues. The last domain explores the responsibility the physician has to the
community, and emphasises the formation of relationships centred around trust.
Heard et al (51) in their study explored 17 factors that may encourage anaesthetists to report
errors. The statement with the highest agreement was found to be “generalised de-identified
feedback about reports received from the anaesthetic community”, with nearly 95% of the
respondents in agreement. This response was echoed in studies by other authors (33, 55,
56). A second factor that predominated was that of senior members of staff encouraging
reporting, with 91% of respondents agreeing or strongly agreeing.
Flin and Yule (57) conducted a review of the effects of leadership on safety in health care.
The authors found that senior managers have “a prime influence on the organisation’s safety
culture”. Heard at al (51) further described how in a “free-text” section that was present in
the questionnaire, respondents eagerly wrote how support from department heads who were
not judgemental, and senior anaesthesiologists discussing their own errors with junior staff,
promoted reporting of adverse events in their environment.
Heard et al (51) also found that nearly 90% of respondents agreed with the statement that
legislated protection of information provided from use in litigation was important. Other
factors in agreement were the ability to report anonymously (84%), clear guidelines about
which adverse events to report (82.8%), how confidentiality would be kept if the report
supplied their details (79.5%), and individualised feedback regarding the reports submitted
(79%).
Factors that found less support when analysed by Heard et al (51) were continuing
professional development points for reporting (66.2%), the ability to report at home by
computer (65%), education about the purpose of reporting (58,7%), computer based
reporting systems (52.3%), and training on how to fill in forms (approx. 50%). The statement
that had the least support was that of payment received for time taken to report, with only
20% of respondents agreeing.
20
In summary, even though there are well established adverse event and error reporting
systems that have been instituted, evidence shows that there are still significant rates of
under-reporting. Some studies have attempted to understand the barriers to reporting of
adverse events and errors, but none have done so in the context of a developing nation
such as South Africa, with its own unique health care challenges
“The currency of patient safety can only be measured in terms of harm prevented and lives
saved. It is the vision of the World Alliance that effective patient safety reporting systems will
help to make this a reality for future patients worldwide.” (19)
21
1.8 References
1. Institute of Medicine Committee on Quality of Health Care in A. In: Kohn LT, Corrigan JM, Donaldson MS, editors. To Err is Human: Building a Safer Health System. Washington (DC): National Academies Press (US)
Copyright 2000 by the National Academy of Sciences. All rights reserved.; 2000.
2. Gaba DM. Anaesthesiology as a model for patient safety in health care. The British Medical Journal. 2000;320(7237):785-8.
3. Beecher HK, Todd DP. A study of the deaths associated with anesthesia and surgery: based on a study of 599, 548 anesthesias in ten institutions 1948-1952, inclusive. Annals of Surgery. 1954;140(1):2-35.
4. Haller G, Laroche T, Clergue F. Morbidity in anaesthesia: today and tomorrow. Best Practice & Research Clinical Anaesthesiology. 2011;25(2):123-32.
5. Smith A, Goodwin D, Mort M, Pope C. Adverse events in anaesthetic practice: qualitative study of definition, discussion and reporting. British Journal of Anaesthesia. 2006;96(9):715-21.
6. Gordon PC, Llewellyn RL, James MF. Drug administration errors by South African anaesthetists--a survey. South African Medical Journal. 2006;96(7):630-2.
7. Botney R. Improving patient safety in anesthesia: a success story? International Journal of Radiation Oncology, Biology, Physics. 2008;71(1 Suppl):S182-6.
8. Runciman W, Merry A. The Wondrous Story of Anesthesia. Eger E, editor. New York: Springer; 2014.
9. Flanagan JC. The critical incident technique. Psychological Bulletin. 1954;51(4):327-58.
10. Cooper JB, Newbower RS, Long CD, McPeek B. Preventable anesthesia mishaps: a study of human factors. Anesthesiology. 1978;49(6):399-406.
11. Williamson JA, Webb RK, Pryor GL. Anaesthesia safety and the "critical incident" technique. Australian clinical review / Australian Medical Association [and] the Australian Council on Hospital Standards. 1985(17):57-61.
12. Australian Patient Safety Foundation [Available from: http://www.apsf.net.au/history.php.
13. National Patient Safety Agency [cited 2014, 21 October]. Available from: https://report.nrls.nhs.uk/nrlsreporting/.
14. Staender S, Davies J, Helmreich B, Sexton B, Kaufmann M. The anesthesia critical incident reporting system: an experience based database. Int J Med Inform. 1997;47:87-90.
15. Chang A, Schyve PM, Croteau RJ, O'Leary DS, Loeb JM. The JCAHO patient safety event taxonomy: a standardized terminology and classification schema for near misses and
22
adverse events. International Journal for Quality in Health Care : Journal of the International Society for Quality in Health Care / ISQua. 2005;17(2):95-105.
16. Tamuz M, Thomas EJ, Franchois KE. Defining and classifying medical error: lessons for patient safety reporting systems. Quality & Safety in Health Care. 2004;13(1):13-20.
17. Staender S. Incident reporting in anaesthesiology. Best Practice & Research Clinical Anaesthesiology. 2011;25(2):207-14.
18. Walshe K. Adverse events in health care: issues in measurement. Quality in Health Care : QHC. 2000;9(1):47-52.
19. WHO Draft Guidelines for Adverse Event Reporting and Learning Systems. Geneva, Switzerland: World Health Organization, 2005.
20. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. The New England Journal of Medicine. 1991;324(6):370-6.
21. Leape LL, Lawthers AG, Brennan TA, Johnson WG. Preventing medical injury. Quality Review Bulletin. 1993;19(5):144-9.
22. Wilson RM, Runciman WB, Gibberd RW, Harrison BT, Newby L, Hamilton JD. The Quality in Australian Health Care Study. The Medical Journal of Australia. 1995;163(9):458-71.
23. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospital patients. The Journal of the American Medical Association. 1991;266(20):2847-51.
24. Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. The Journal of the American Medical Association. 1995;274(1):29-34.
25. Andrews LB, Stocking C, Krizek T, Gottlieb L, Krizek C, Vargish T, et al. An alternative strategy for studying adverse events in medical care. Lancet. 1997;349(9048):309-13.
26. Starmer AJ, Sectish TC, Simon DW, Keohane C, McSweeney ME, Chung EY, et al. Rates of medical errors and preventable adverse events among hospitalized children following implementation of a resident handoff bundle. The Journal of the American Medical Association. 2013;310(21):2262-70.
27. Catchpole K, Bell MD, Johnson S. Safety in anaesthesia: a study of 12,606 reported incidents from the UK National Reporting and Learning System. Anaesthesia. 2008;63(4):340-6.
28. Gupta S, Naithani U, Brajesh SK, Pathania VS, Gupta A. Critical incident reporting in anaesthesia: a prospective internal audit. Indian Journal of Anaesthesia. 2009;53(4):425-33.
29. Lundgren AC. Peri-Operative death in two major academic hospitals in Johannesburg, South Africa. [Unpublished Thesis]. In press 2011.
23
30. Madzimbamuto FD, Chiware R. A critical incident reporting system in anaesthesia. The Central African Journal of Medicine. 2001;47(11-12):243-7.
31. Labuschagne M, Robbetze W, Rozmiarek J, Strydom M, Wentzel M, Diedericks BJ, et al. Errors in drug administration by anaesthetists in public hospitals in the Free State. South African Medical Journal. 2011;101(5):324-7.
32. Holden RJ, Karsh BT. A review of medical error reporting system design considerations and a proposed cross-level systems research framework. Human Factors. 2007;49(2):257-76.
33. Pham JC, Girard T, Pronovost PJ. What to do with healthcare incident reporting systems. Journal of Public Health Research. 2013;2(3):e27.
34. Suresh G, Horbar JD, Plsek P, Gray J, Edwards WH, Shiono PH, et al. Voluntary anonymous reporting of medical errors for neonatal intensive care. Pediatrics. 2004;113(6):1609-18.
35. Leape LL. Reporting of adverse events. The New England Journal of Medicine. 2002;347(20):1633-8.
36. Billings CE. Some hopes and concerns regarding medical event-reporting systems: lessons from the NASA Aviation Safety Reporting System. Archives of Pathology & Laboratory Medicine. 1998;122(3):214-5.
37. Runciman WB. Qualitative versus quantitative research--balancing cost, yield and feasibility. 1993. Quality & Safety in Health Care. 2002;11(2):146-7.
38. Smith ML, Forster HP. Morally managing medical mistakes. Cambridge quarterly of healthcare ethics : CQ : The International Journal of Healthcare Ethics Committees. 2000;9(1):38-53.
39. Evans SM, Berry JG, Smith BJ, Esterman A, Selim P, O'Shaughnessy J, et al. Attitudes and barriers to incident reporting: a collaborative hospital study. Quality & Safety in Health Care. 2006;15(1):39-43.
40. Wood KE, Nash DB. Mandatory state-based error-reporting systems: current and future prospects. American journal of medical quality : the official journal of the American College of Medical Quality. 2005;20(6):297-303.
41. Williams SK, Osborn SS. The development of the National Reporting and Learning System in England and Wales, 2001-2005. The Medical journal of Australia. 2006;184(10 Suppl):S65-8.
42. MacLeod H. Solutions to problems that have already been solved. [updated 4 February, 201321 October, 2014]. Available from: http://hospitalnews.com/solutions-to-problems-that-have-already-been-solved/.
43. Barach P, Small SD. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting systems. The British Medical Journal. 2000;320(7237):759-63.
44. Kaplan H, Barach P. Incident reporting: science or protoscience? Ten years later. Quality & Safety in Health Care. 2002;11(2):144-5.
24
45. Vincent CA. Analysis of clinical incidents: a window on the system not a search for root causes. Quality & Safety in Health Care. 2004;13(4):242-3.
46. Henriksen K, Kaplan H. Hindsight bias, outcome knowledge and adaptive learning. Quality & Safety in Health Care. 2003;12 Suppl 2:ii46-50.
47. Taylor JA, Brownstein D, Christakis DA, Blackburn S, Strandjord TP, Klein EJ, et al. Use of incident reports by physicians and nurses to document medical errors in pediatric patients. Pediatrics. 2004;114(3):729-35.
48. Braithwaite J, Westbrook M, Travaglia J. Attitudes toward the large-scale implementation of an incident reporting system. International Journal for Quality in Health Care. 2008;20(3):184-91.
49. Woolf SH, Kuzel AJ, Dovey SM, Phillips RL, Jr. A string of mistakes: the importance of cascade analysis in describing, counting, and preventing medical errors. Annals of Family Medicine. 2004;2(4):317-26.
50. Jeffe DB, Dunagan WC, Garbutt J, Burroughs TE, Gallagher TH, Hill PR, et al. Using focus groups to understand physicians' and nurses' perspectives on error reporting in hospitals. Joint Commission Journal on Quality and Safety. 2004;30(9):471-9.
51. Heard GC, Sanderson PM, Thomas RD. Barriers to adverse event and error reporting in anesthesia. Anesthesia and Analgesia. 2012;114(3):604-14.
52. Kaldjian LC, Jones EW, Rosenthal GE. Facilitating and impeding factors for physicians' error disclosure: a structured literature review. Joint Commission Journal on Quality and Patient Safety / Joint Commission Resources. 2006;32(4):188-98.
53. Kaldjian LC, Jones EW, Rosenthal GE, Tripp-Reimer T, Hillis SL. An empirically derived taxonomy of factors affecting physicians' willingness to disclose medical errors. Journal of General Internal Medicine. 2006;21(9):942-8.
54. Wu AW. Medical error: the second victim. The doctor who makes the mistake needs help too. The British Medical Journal. 2000;320(7237):726-7.
55. Elder NC, Graham D, Brandt E, Hickner J. Barriers and motivators for making error reports from family medicine offices: a report from the American Academy of Family Physicians National Research Network (AAFP NRN). Journal of the American Board of Family Medicine : JABFM. 2007;20(2):115-23.
56. Walker P, Pekmezaris R, Lesser ML, Nouryan CN, Rosinia F, Pratt K, et al. A multisite validity study of self-reported anesthesia outcomes. American Journal of Medical Quality : the official journal of the American College of Medical Quality. 2012;27(5):417-25.
57. Flin R, Yule S. Leadership for safety: industrial experience. Quality & Safety in Health Care. 2004;13 Suppl 2:ii45-51.
25
SECTION 2 – Journal guidelines to authors
This section highlight’s the guidelines which the author has followed with regards to the
length and formatting of the research article.
The guidelines followed in creating the draft article were those of the Southern African
Journal of Anaesthesia and Analgesia, which is the intended journal of publication.
26
Southern African Journal of Anaesthesia and Analgesia guidelines to authors
Article sections and length
The following contributions are accepted (word counts exclude abstracts, tables and
references):
Original research (2800 – 3200 words/ 4-5 pages)
FULL AUTHOR GUIDELINES
Title page
All articles must have a title page with the following information and in this particular order:
Title of the article; surname, initials, qualifications and affiliation of each author; The name,
postal address, e-mail address and telephonic contact details of the corresponding author
and at least 5 keywords.
Abstract All articles should include an abstract. The structured abstract for an Original Research
article should be between 200 and 230 words and should consist of four paragraphs labeled
Background, Methods, Results, and Conclusions. It should briefly describe the problem or
issue being addressed in the study, how the study was performed, the major results, and
what the authors conclude from these results. The abstracts for other types of articles should
be no longer than 230 words and need not follow the structured abstract format.
Keywords
All articles should include keywords. Up to five words or short phrases should be used. Use
terms from the Medical Subject Headings (MeSH) of Index Medicus when available and
appropriate. Key words are used to index the article and may be published with the abstract.
Acknowledgements
In a separate section, acknowledge any financial support received or possible conflict of
interest. This section may also be used to acknowledge substantial contributions to the
research or preparation of the manuscript made by persons other than the authors.
References
Cite references in numerical order in the text, in superscript format (Format> Font> Click
superscript). Please do not use brackets or do not use the foot note function of MS Word.
27
In the References section, references must be typed double-spaced and numbered
consecutively in the order in which they are cited, not alphabetically.
The style for references should follow the format set forth in the Uniform Requirements for
Manuscripts Submitted to Biomedical Journals (http://www.icmje.org) prepared by the
International Committee of Medical Journal Editors. Abbreviations for journal titles should
follow Index Medicus format. Authors are responsible for the accuracy of all references.
Personal communications and unpublished data should not be referenced. If essential, such
material should be incorporated in the appropriate place in the text.
List all authors when there are six or fewer; when there are seven or more, list the first three,
then ";et al."; When citing URLs to web documents, place in the reference list, and use the
following format: Authors of document (if available). Title of document (if available). URL.
(Accessed [date]).
The following are sample references:
1. Jun BC, Song SW, Park CS, Lee DH, Cho KJ, Cho JH. The analysis of
maxillary sinus aeration according to aging process: volume assessment by 3-
dimensional reconstruction by high-resolutional CT scanning. Otolaryngol Head Neck
Surg. 2005 Mar;132(3):429-34.
2. 2. Polgreen PM, Diekema DJ, Vandeberg J, Wiblin RT, Chen YY, David S, et
al. Risk factors for groin wound infection after femoral artery catheterization: a case-
control study. Infect Control Hosp Epidemiol [Internet]. 2006 Jan [cited 2007 Jan
5];27(1):34-7. Available
from: http://www.journals.uchicago.edu/ICHE/journal/issues/v27n1/2004069/2004069
.web.pdf.
More sample references can be found
at: http://www.nlm.nih.gov/bsd/uniform_requirements.html"
Tables
Tables should be self-explanatory, clearly organised, and supplemental to the text of the
manuscript. Each table should include a clear descriptive title on top and numbered in
Roman numerals (I, II, etc) in order of its appearance as called out in text. Tables must be
inserted in the correct position in the text. Authors should place explanatory matter in
footnotes, not in the heading. Explain in footnotes all non-standard abbreviations.
For footnotes use the following symbols, in sequence: *, †, ‡, §, ||, **, ††, ‡‡
28
Figures
All figures must be inserted in the appropriate position of the electronic document. Symbols,
lettering, and numbering (in Arabic numerals e.g. 1, 2, etc. in order of appearance in the
text) should be placed below the figure, clear and large enough to remain legible after the
figure has been reduced. Figures must have clear descriptive titles.
Photographs and images
If photographs of patients are used, either the subject should not be identifiable or use of the
picture should be authorised by an enclosed written permission from the subject. The
position of photographs and images should be clearly indicated in the text. Electronic images
should be saved as either jpeg or gif files. All photographs should be scanned at a high
resolution (300dpi, print optimised). Please number the images appropriately.
Permission
Permission should be obtained from the author and publisher for the use of quotes,
illustrations, tables, and other materials taken from previously published works, which are
not in the public domain. The author is responsible for the payment of any copyright fee(s) if
these have not been waived. The letters of permission should accompany the manuscript.
The original source(s) should be mentioned in the figure legend or as a footnote to a table.
Review and action
Manuscripts are initially examined by the editorial staff and are usually sent to independent
reviewers who are not informed of the identity of the author(s). When publication in its
original form is not recommended, the reviewers' comments (without the identity of the
reviewer being disclosed) may be passed to the first author and may include suggested
revisions. Manuscripts not approved for publication will not be returned.
Ethical considerations
Papers based on original research must adhere to the Declaration of Helsinki on ";Ethical
Principles for Medical Research Involving Human Subjects"; and must specify from which
recognised ethics committee approval for the research was obtained.
Conflict of interest Authors must declare all financial contributions to their work or other forms of conflict of
interest, which may prevent them from executing and publishing unbiased research. [Conflict
of interest exists when an author (or the author's institution), has financial or personal
relationships with other persons or organizations that inappropriately influence (bias) his or
her opinions or actions.]* *Modified from: Davidoff F, et al. Sponsorship, Authorship, and
29
Accountability. (Editorial) JAMA 2001: 286(10) The following declaration may be used if
appropriate: "I declare that I have no financial or personal relationship(s) which may have
inappropriately influenced me in writing this paper."
Submissions and correspondence
All submissions must be made online at www.sajaa.co.za and correspondence regarding
manuscripts should be addressed to:
The Editor, SAJAA, E-mail: [email protected]
Note: Ensure that the article ID [reference] number is included in the subject of your email
correspondence.
Electronic submissions by post or via email Authors with no e-mail or internet connection can mail their submissions on a CD to: SAJAA,
PO Box 14804, Lyttelton Manor, 0140, Gauteng, South Africa.
All manuscripts will be processed online. Submissions by post or by e-mail must be
accompanied by a signed copy of the following indemnity and copyright form. CLICK
HERE to download and save it to your computer.
30
Tips on Preparing your manuscript
1. Please consult the “Uniform requirements for manuscripts submitted to
biomedical journals” at www.icmje.org 2. Please consult the guide on Vancouver referencing methods
at: http://www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=citmed.TOC&depth=2
3. The submission must be in UK English, typed in Microsoft Word or RTF
with no double spaces after the full stops, double paragraph spacing, font size 10
and font type: Times New Roman.
4. All author details (Full names, Qualifications and affiliation) must be
provided.
5. The full contact details of corresponding author (Tel, fax, e-mail, postal
address) must be on the manuscript.
6. There must be an abstract and keywords.
7. References must strictly be in Vancouver format. (Reference numbers must
be strictly numerical and be typed in superscript, not be in brackets and must be
placed AFTER the full stop or comma.)
8. It must be clear where every figure and table should be placed in the text. If
possible, tables and figures must be placed in the text where appropriate. If too
large or impractical, they may be featured at the end of the manuscript or uploaded
as separate supplementary files.
9. All photographs must be at 300dpi and clearly marked according to the
figure numbers in the text. (Figure 1, Table II, etc.)
10. All numbers below ten, without percentages or units, must be written in
words.
11. Figure numbers: Arabic, table numbers: Roman
<="" uk="" in="" proficient="" expert="" a="" by="" reviewed="" be="" must="">
31
SECTION 3 – Draft Article
Factors influencing adverse event and error reporting in Anaesthesiology
Authors
Steven Nel, MBBCh (Witwatersrand), DA (SA)
Juan Scribante, M Cur
Helen Perrie, MSc
Christina Lundgren, FFA (SA); PhD; MSc (Med) Health Law and Bioethics
Department of Anaesthesiology, School of Clinical Medicine, Faculty of Health Sciences, University of the
Witwatersrand.
Corresponding author:
Nel, S.
Department of Anaesthesiology, Chris Hani Baragwanath Academic Hospital, 26 Chris Hani Road,
Johannesburg, 1860
+2711 9339335
Keywords:
Anaesthesia, critical incident, adverse event, reporting systems, patient safety.
32
Abstract
Background
Adverse events and errors are a widespread cause of morbidity and mortality in the health care environment.
Adverse event and error reporting systems have been shown to potentially reduce the occurrence of these
events, however there is still significant under-reporting. Little is known regarding the barriers to reporting of
adverse events and errors in the context of South Africa, or what emotional and attitudinal barriers may be
present regarding a formal reporting system amongst anaesthetists in the Department of Anaesthesiology at the
University of the Witwatersrand.
Methods
A prospective, descriptive, contextual study design utilizing an anonymous self-administered questionnaire was
distributed to 133 anaesthetists who attended academic anaesthetic meetings.
Results
One hundred and eighteen questionnaires met the criteria for analysis, giving a response rate of 92%. Barriers to
reporting included a “code of silence” in medicine and blame from colleagues. If a specified error as opposed to
an adverse event had occurred, participants were more likely to agree with barriers regarding fear of litigation,
disciplinary action, getting into trouble, as well as colleagues that may be unsupportive. Strategies to promote
reporting of adverse events and errors include senior role models who encourage reporting and individualised
feedback regarding reports made.
Conclusions
Most anaesthetists in our study disagreed with barriers to reporting an unspecified adverse event. However, if an
error has occurred, reporting behaviour may be inhibited by barriers regarding fears of litigation, disciplinary
action and lack of support. Senior role models that openly support reporting along with individualised feedback
may increase reporting rates.
33
Introduction
Adverse events and errors are still a widespread cause of morbidity and mortality in the health care
environment,1-5 with ample literature indicating that many of these errors are often preventable.5-9 Adverse
events and errors have been increasingly recognised and focused upon since the report released in 1999 by the
Institute of Medicine titled “To Err is Human: Building a safer health system”. The report highlighted that there
was an “epidemic of medical errors” occurring in health care, and advocated for the provision of safer care for
patients. It showed that between 44,000 to 98,000 patients die in hospitals in the United States annually due to
medical errors that were potentially preventable.10 Rates of adverse events or errors vary in the literature, but
range from approximately 16% to almost 46% of patients admitted. Serious adverse event rates were found to be
as high as 17% of admissions.2, 11
A key trend within the medical community currently is to maximise patient safety and reduce medical errors or
harm to patients.10 The field of anaesthesia is known to be a leader in good safety practices, having initially
pioneered adverse event reporting in the medical domain, using lessons learnt from the aviation industry12 as
well as having shown a considerable decline in mortality rates over the last 50 years.13-17 In an effort to reduce
adverse events, there has been research into improving patient safety by exploring any factors that may
potentially lead to harm. Evidence supports the fact that a large percentage of adverse events are from “system
failures” and “organisational factors”, which are potentially avoidable.18 As a result of the high rate of
preventable errors, many medical institutions and societies have instituted formal adverse event and error
reporting systems as a way to improve patient safety.19
Benefits of such systems are that analyses can be performed on various incidents or adverse events, which can
be used for further learning, and to facilitate the framework for robust prevention strategies, possibly leading to
reduced adverse events in the future. It also creates a societal culture of reporting, which is currently lacking
within the health care field when compared with other critical environments or industries.16, 20, 21
Even though there are institutions that have adverse event reporting systems in place, there is still a significant
problem with under-reporting of adverse events and errors.16, 22 Reasons for this include factors such as time
available to report, lack of anonymity, lack of clarity of definitions of adverse events, and perceived seriousness
of the event itself. Some observational studies have shown that up to 90% of adverse events and errors are not
being reported.20, 23
34
Heard et al,23 in 2012, identified barriers to adverse event and error reporting amongst anaesthetists in Australia
and New Zealand, and described factors which may improve reporting. The authors found that a significant
barrier to reporting was fear of blame by colleagues. They also described further barriers to reporting when
doctors had made an error that lead to an adverse event as compared to doctors who had not made an error even
though an adverse event had occurred. There is little research describing barriers to reporting of adverse events
and errors in the context of a developing nation such as South Africa, especially where adverse event reporting
systems are not well established and seldom used. This study aimed to describe perceived barriers to the
reporting of adverse events and errors amongst anaesthetists in the Department of Anaesthesiology at the
University of the Witwatersrand, as well as to propose factors that may promote or encourage the reporting of
adverse events and errors.
Method
Approval to conduct this study was obtained from the Human Research Ethics Committee (Medical) of the
University of the Witwatersrand, and other relevant authorities. This was a prospective, contextual, descriptive
study where anaesthetists in the Department of Anaesthesiology were asked to voluntarily complete an
anonymous self-administered questionnaire at departmental academic meetings. Consent was implied on
completion of the questionnaire, which was placed in a sealed box. One author (SN) was available to answer
questions.
The questionnaire used was modified slightly from that published by Heard et al,23 who derived their
questionnaire through an extensive review of the available literature. The conduct and content of this
questionnaire varied from that by Heard et al23 in the following aspects, namely, questionnaires were completed
by anaesthetists at departmental academic meetings rather than by mailed surveys, and the statement “ANZCA
Continuing Professional Development points for reports” was modified to “SASA Continuing Professional
Development points for reports” to make it appropriate for a South African context. The research design also
varied, as Heard et al23 used a randomised between-groups design, whereas this study utilised a single-subject
design for the scenario questions. The questionnaire consisted of five sections, with three sections listing
statements to be rated using a 5-point Likert Scale. Possible selections ranged from “strongly agree” to “strongly
disagree”. The layout of the questionnaire is shown in Table I.
35
Table I: Summary of questionnaire structure
Section Intention Description
Section 1 Demographic characteristics Single answer selection from a range of variables
Knowledge of existing adverse event and
error reporting systems
Section 2 Explore attitudinal and emotional factors
regarding the reporting of adverse events
13 statements on a 5-point Likert scale. 10
statements focused on attitudinal barriers, and 3
statement focused on fears and anxiety barriers
Section 3 Determine barriers to the reporting of an
adverse event (section 3A)
17 statements on a 5-point Likert scale, with a
case scenario where a patient has an anaphylactic
reaction to an antibiotic administered, with no
pre-operative history of any allergy (an adverse
event)
Determine barriers to the reporting of an
error (section 3B)
17 statements on a 5-point Likert scale, with a
case scenario where a patient has an anaphylactic
reaction to an antibiotic administered, with the
anaesthetist recalling a pre-operative history of an
allergy to that specific antibiotic (an error)
Section 4 Strategies that may promote or improve the
reporting of adverse events and errors
17 statements on a 5-point Likert scale
Section 5 Further comments on current reporting
structures, further barriers to reporting, and
factors that may facilitate reporting of events
Free text paragraph
Of the 208 anaesthetists eligible to participate, it was estimated that 166 (80%) would be available at academic
meetings, with the remainder involved with other commitments such as annual leave and emergency calls. Of
the potential 166 participants, a response rate of 80% was targeted (n = 133). Interns rotating through
anaesthesia were excluded from participating. Returned questionnaires that were less than 50% complete were
used to calculate the response rate, but not included in data analysis.
36
Data were captured onto spreadsheets using Microsoft Excel® 2011. GraphPad Prism® v5.02 was used to
analyse data, in consultation with a biostatistician. Categorical variables were presented using frequencies and
percentages. For section 3 the Likert scale data were treated as interval scale data, with a score of 1 assigned to
strongly agree, successively to a score of 5 for strongly disagree, thus a higher score showing higher levels of
disagreement. Wilcoxon matched pairs were then used to compare the responses between the two groups. A P
value of <0.05 was considered to be statistically significant. As per Heard et al,23 the Likert scales used in
sections 2, 3 and 4 were converted from 5-point to 3-point Likert scales, i.e. agree and strongly agree to agree,
neutral remained the same, and disagree and strongly disagree to disagree. This was presented as ordinal scale
data. Percentages were rounded off one decimal point.
Results
One hundred and thirty-three questionnaires were distributed at four consecutive departmental academic
meetings, with a total of 123 returned, yielding a response rate of 74%. Five questionnaires were excluded as
they were less than 50% complete (n = 118). Twenty-four questionnaires had some data missing, however they
were all included in the data analysis. Of these, two participants did not complete section 3B at all, and were
thus excluded from the Wilcoxon matched pairs analysis (n = 116). Table II details participants’ demographic
data.
37
Table II: Participant’s demographic data
Characteristic Frequency (n) Percentage (%) Gender Male 49 41.5 Female 68 57.6 Unknown 1 0.9 Age (years) 21 - 30 26 22.1 31 - 40 64 54.2 41 - 50 8 6.8 51 - 60 15 12.7 61 - 70 3 2.5 Unknown 2 1.7 Professional designation Medical officer 25 21.2 Registrar 47 39.8 Consultant 36 30.5 Unknown 10 8.5 Years of experience Less than 5 53 44.9 5 - 10 40 33.9 11 - 20 10 8.5 > 20 14 11.9 Unknown 1 0.8
Table III shows participants knowledge of current reporting structures available at their workplace, with 105
(89%) participants aware of a reporting system at their workplace, and 94 (90%) participants indicating that the
reporting system available was a formal system that was paper or computer based.
Table III: Current reporting structures available
Characteristic Frequency (n) Percentage (%) Adverse event or error reporting system at workplace? Yes 105 89.0 No 3 2.5 Do not know 10 8.5 If yes, formal or Informal system? Formal system 94 89.5 Informal system 11 10.5 If yes, who instituted it? Anaesthesiology department 88 83.8 Hospital 2 1.9 Provincial government 1 1 National government 0 0 Do not know 14 13.3
38
Figure 1 shows the results from Section 2 of the questionnaire, which explored participants’ attitudes and
emotional factors regarding the reporting of adverse events and errors. For 11 out of 13 statements most
participants were in disagreement. The only statement with which more participants agreed than disagreed was
“Medicine has a culture of silence where errors are not talked about”, with only 32% disagreeing. A second
statement “Doctors who make errors are blamed by their colleagues” had the same levels of agreement versus
disagreement (34%). The two statements with the highest levels of disagreement were “Competition with my
peers would prevent me from disclosing an error” and “I would cover up an error if I could”.
Figure 1: Attitudinal and emotional factors regarding the reporting of adverse events
5
7
11
14
10
15
5
14
27
23
50
34
12
12
3
13
11
17
15
9
14
13
22
18
32
20
83
90
76
75
72
70
85
73
60
54
32
34
68
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Iwouldprotectmyself-interestsaheadoftheinterestsofthepatientifIcould,e.g.,byhidingordenyinganerror
Competitionwithmypeerswouldpreventmefromdisclosinganerror
Ifadoctoriscarefulenoughheorshewillnotmakeanerror
Itwouldaffectmyidentityasadoctortoadmittoanerror
Othersdon’tneedtoknowabouterrorsIhavemade
Disclosinganerror,ifyoudon’thaveto,isanoptionalactofheroism
IwouldcoverupanerrorIhadmadeifIcould
IfIadmittoanerror,Iwillfeellikeafailure
Itwouldaffectmyself-esteemtoadmittoanerror
Doctorswhomakeerrorsarehumiliatedbytheircolleagues
Medicinehasacultureofsilencewhereerrorsarenottalkedabout
Doctorswhomakeerrorsareblamedbytheircolleagues
Doctorsshouldnotmakeerrors
Agree Neutral Disagree
39
Section 3 explored potential barriers to the reporting of an adverse event or error with a given clinical scenario.
When comparing participant’s responses for section 3A where no error had occurred as opposed to section 3B
where an error had occurred, participants were more likely to agree with four statements when an error had been
made, all with p-values of < 0.0001 (Table IV). These statements were:
• “I am worried about litigation” (76% vs 31%)
• “I don’t want to get into trouble” (63% vs 45%)
• “My colleagues may be unsupportive” (47% vs 13%)
• “I am worried about disciplinary action” (59% vs 13%)
For the above statements, if an adverse event had occurred rather than an error, participants were more likely to
disagree with those statements as potential barriers. With regards to the other 13 potential barriers to reporting,
participants were more likely to disagree with the statements, regardless of whether the scenario was associated
with an adverse event or an error. Of these 13 statements, only seven found statistical significance.
Where the median values and interquartile ranges were the same in a given question, the sum of ranks showed
where the difference occurred. A lower sum of rank indicated a greater level of agreement and a higher sum of
rank indicated a greater level of disagreement.
40
Table IV: Potential barriers to reporting an adverse event or error when anaphylaxis occurred in a given
clinical scenario*
* In the clinical scenario in section 3A there was an adverse event but no error, whereas in section 3B there
was an adverse event as a result of an error in giving an antibiotic to a patient with a known allergy to that
antibiotic, causing anaphylaxis. NE = No-error, E = Error
Statement Group Sum of Rank
Median (Interquartile
Range) P value
1. I am worried about litigation. NE 391 4 (2-4) <0.0001
E 257 2 (2-2) 2. I don’t want to get into trouble. NE 359 3 (2-4)
<0.0001 E 299 2 (2-3)
3. My colleagues may be unsupportive. NE 444 4 (3-4) <0.0001
E 331 3 (2-4) 4. I am worried about disciplinary action. NE 447 4 (3-4)
<0.0001 E 307 2 (2-4)
5. I may be blamed unfairly for the event. NE 423 4 (3-4) 0.001
E 376 4 (2-4) 6. I do not want the case discussed at meetings. NE 457 4 (4-5)
<0.0001 E 386 4 (2-4)
7. Adverse event reporting makes little contribution to quality of care.
NE 513 5 (4-5) 0.0029
E 489 4 (4-5) 8. I don’t know whose responsibility it is to make
a report. NE 476 4 (4-5)
0.2061 E 485 4 (4-5)
9. A good outcome of the case makes reporting unnecessary
NE 481 4 (4-5) 0.2061
E 490 4 (4-5) 10. I don’t know which adverse events should be
reported. NE 420 4 (3-4)
0.0117 E 441 4 (3-4)
11. Even if I don’t give details I am worried they will track me down.
NE 463 4 (4-5) 0.0081
E 440 4 (3-4) 12. The forms take too long to fill in and I just don’t
have time. NE 402 4 (3-4)
0.0079 E 419 4 (3-4)
13. When I am busy at work I forget to make a report.
NE 368 3 (2-4) 0.0167
E 383 4 (2-4) 14. I don’t feel confident the information will be
kept confidential. NE 402 4 (3-4)
0.2691 E 393 4 (3-4)
15. I never get any feedback after I report an adverse event.
NE 362 3 (2-4) 0.3216
E 357 3 (2-4) 16. I wonder about who else will have access to the
information. NE 369 3 (2-4)
0.0837 E 358 3 (2-4)
17. As long as the staff involved learn from incidents it is unnecessary to discuss any further.
NE 469 4 (4-5) 0.9723
E 469 4 (4-5)
41
In Section 4 strategies that may promote or improve reporting of adverse events and errors were investigated
(Figure 2). The statement with the most agreement (98%) was “Role models, e.g. senior colleagues who openly
encourage reporting”. There were high levels of agreement for most statements. The statement that had the
highest level of disagreement was “payment taken for time to report” with 62% disagreeing. The statement
“SASA Continuing Professional Development points for reports” also had less support with 55% agreeing.
Figure2:Strategiesthatmaypromoteorimprovethereportingofadverseeventsanderrors
83
98
82
86
91
86
92
71
82
86
55
86
85
89
79
68
21
10
14
5
4
9
3
13
14
12
23
8
9
6
16
24
17
6
2
3
9
4
5
4
16
3
3
22
6
5
5
5
8
62
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Generalizedde-identifiedfeedbackaboutreportsreceivedfromtheanaestheticcommunity
Rolemodels,e.g.seniorcolleagueswhoopenlyencouragereporting
Legislatedprotectionofinformationyouprovidefromuseinlitigation
Abilitytoreportanonymously
Clearguidelinesaboutwhatadverseeventsanderrorstoreport
Informationonhowconfidentialitywillbemaintainedifyousupplyyourname
Individualizedfeedbacktoyouaboutreportsyousubmit
Paperformsforreportingineachtheatre
Moresupportfromcolleagues
Lessblameattachedtothosewhoreporterrors
SASAContinuingProfessionalDevelopmentpointsforreports
Accesstocomputer-basedreportingsystemsfromhome
Educationaboutthepurposeofreporting
Computer-basedreportingsystems
Trainingonhowtousecomputer-basedreportingsystems
Trainingonhowtofillinpaperformsforreporting
Paymentfortimetakentoreport
Agree Neutral Disagree
42
Section 5 was for participant’s free text, and allowed for any comments regarding adverse event and error
reporting structures, as well as any further factors that may influence reporting. Only 37 (31%) participants
completed the comments section. Common themes with regards to barriers to reporting included fear of
litigation, discrimination and blame from colleagues. Other barriers included not knowing what or when to
report, and forgetting to report. Factors suggested that may promote or enhance reporting included support from
senior colleagues, individualised feedback and debriefing, making reporting easier with shorter forms and
electronic submissions, and having a clear reporting structure in place. Many participants felt it should be
compulsory to report, with some stating it should be a legal requirement.
Discussion
Of the 118 participants in our study, there was a female predominance (58%), with 76% of participants falling
between the ages of 21 to 40 years. Most participants (89%) were aware of some form of adverse event or error
reporting system at their workplace, with 84% stating that the anaesthetics department at their workplace had
instituted the reporting system.
Section 2 explored anaesthetist’s attitudes to the reporting of adverse events and errors. The statement that had
the most agreement amongst participants (50%) was “Medicine has a culture of silence where errors are not
talked about”. This contrasted with the 37% who agreed with the same statement by Heard et al’s study23 This
confirms that a “code of silence” is an important factor that may inhibit the reporting of adverse events, a
concern also expressed by other authors.24, 25 A second barrier to reporting that was prevalent was that of
“blame”, with 34% of participants agreeing that colleagues will blame them for errors made, which is lower
than the 46% who agree in Heard et al’s study23 This belief is strongly supported in the literature, and is
associated with serious consequences such as anxiety, depression and suicide.21, 24, 26-28 If analysis of adverse
events and errors only focuses on the individual involved, and directs blame, then deeper and more meaningful
analysis of system error and root cause analysis will not be possible, seriously limiting the usefulness of
reporting systems as a learning and patient safety tool.27, 29-32 Excluding the barriers of blame and a culture of
silence, participants in our study generally disagreed with most of the proposed barriers to reporting.
Importantly, between 68 to 76% of participants disagreed with the statements “Doctors should not make errors”
and “If a doctor is careful he or she will not make an error”. This showed a trend away from the traditional
thinking of the “medical perfectibility model” where doctors have unrealistic expectations of providing error-
free medical care, towards more realistic expectations that errors can and will happen.24 Only 5% of participants
43
agreed that they would cover up an error if they could, compared with 10% agreement in Heard et al’s study.23
Only 5% of participants in both studies agreed that they would protect their self-interests ahead of those of a
patient’s. Such low levels of agreement may be accurate, or may represent a socially acceptable response of
what “should be answered”. Of concern was that 23% of participants in our study felt that they would be
humiliated by their colleagues when making errors, as compared to 18% by Heard et al,23 signifying a
potentially serious disincentive towards reporting adverse events or errors.
With the given clinical scenario, if participants had made an error as opposed to an adverse event, they were
significantly more likely to agree with barriers concerning fear of litigation (76 vs 31%), not wanting to get into
trouble (63 vs 45%), unsupportive colleagues (47 vs 13%), fear of disciplinary action (59 vs 13%), fear of being
blamed unfairly (30 vs 24%), and not wanting the case to be discussed in meetings (26 vs 8%), with comparable
findings by Heard et al.23 The similar findings between both studies may indicate that these fears or concerns are
universal and legitimate.
Participants’ most significant concern was fear of litigation, with 76% of participants fearing litigation if they
had made an error as opposed to Heard et al23 where there was 58% agreement regarding the same barrier. If
participants had an adverse event as opposed to an error, then only 31% agreed with fear of litigation as a barrier
to reporting. This concern may be explained by an increasing trend towards litigation in South Africa, with high
value claims of greater than R5 million increasing by more than 900% over the last five years.33 This may be
compounded by the fact that adverse event and error reports may be subject to “legal discovery, with very
damaging materials that were intended for safety rather than legal use” being subpoenaed for civil lawsuits.34
Most participants regardless of whether associated with an error or no error scenario disagreed with statements
regarding not knowing which adverse event to report (67-73%), being too busy at work and forgetting to report
(44-53%), forms taking too long to fill and not having the time (55-64%), and that “adverse event reporting
makes little contribution to quality of care” (91-93%). This latter finding is in keeping with those by Heard et
al’s study,23 where approximately 2% agreed that “Adverse events make little contribution to care”.
Disagreement with these statements also showed that time and busyness are not important barriers to reporting,
a finding corroborated by Heard et al’s study,23 but in contrast with findings by other authors.20, 27 The
statement “Even if I don’t give my details I am worried they will track me down” had 73 to 79% disagreement
as compared to the same statement by Heard et al23 which had 62 to 68% disagreement. This differs with other
authors who have found that anonymity influences willingness to report. 26, 35
44
This study found that the most important factor that may facilitate the reporting of adverse events and errors to
be senior colleagues who openly encourage reporting, with 98% of participants agreeing, and none disagreeing.
A similar proportion in agreement of 91% was found by Heard et al.23 This demonstrated how important
leadership can be in creating a successful reporting system. The statement “Individualised feedback to you about
reports you submit” was the second most accepted statement, with 92% agreeing. These statements differ from
Heard et al23 who identified generalised de-identified feedback (94%) rather than individualised feedback (79%)
as the most prevalent statement. With generalised feedback there is potential to learn from previous errors or
adverse events,29, 30, 36 however individualised feedback may allow for debriefing of the event, and gives closure
to the individual involved.37 A third statement where 90% of participants agreed was “clear guidelines about
what adverse events or errors to report”, with 83% of participants in the research by Heard et al23 also agreeing.
Without clear guidelines, errors “can be defined away” by misinterpreting ambiguous definitions, leading to
under-reporting.38 The lack of a universal taxonomy for adverse event and error reporting systems also makes
clear guidelines on what to report important.30, 38-40
The statements with the lowest level of agreement was payment in return for time taken to report (21%), with
similar findings from Heard et al.23 Other statements that had lower levels of agreement were those regarding
continuing professional development points for reporting (55%), which may represent a bias towards registrars
in training who do not need to gain professional development points, and a preference in our study towards
computer based rather than paper based reporting systems (89% vs. 70%). This contrasted with research by
Heard et al,23 where 73% of participants agreed or strongly agreed to paper based reporting, as compared with
only 52% agreement towards computer-based reporting systems.
An important finding from participants’ free text was that many felt that reporting of adverse events and errors
should be compulsory, with some feeling it should even be a legal requirement to report.
Limitations
This study is contextual to the Department of Anaesthesiology at the University of the Witwatersrand, and thus
the results may not be generalisable to other anaesthetic departments or medical departments in South Africa or
other countries. However, an advantage is that results from this study compare closely to those by Heard et al,23
indicating that the findings may be universal to countries with similar medical and legal systems. Responder
bias is a further potential limitation, as participants who completed the questionnaire may be more willing to fill
in adverse event and error forms than those who did not fill in questionnaires.
45
The order of statements and the clinical scenarios utilised may have influenced participants’ responses and
could vary if other scenarios or statements were used, however, this questionnaire was based on that by Heard et
al,23 and strengthens the findings of our study, as our results are comparable to those of Heard et al23. The way
in which participants completed the questionnaire could have been affected by “social desirability bias”, where
participants may have completed the questionnaire in a socially acceptable manner rather than with direct
honesty.
Conclusion
The majority of anaesthetists in our study disagreed with barriers to reporting an unspecified adverse event,
however, if an error has occurred, there is a greater likelihood that reporting behaviour may be inhibited by barriers
regarding fears of litigation, disciplinary action and lack of support. Senior role models that openly support
reporting along with individualised feedback may increase reporting and the associated benefits thereof.
Acknowledgements
This study was in partial fulfilment of the requirements for the degree of Master of Medicine in the branch of
Anaesthesiology.
Conflict of Interest
I declare that I have no financial or personal relationship(s) which may have inappropriately influenced me in
writing this paper.
46
References
1. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. The New England Journal of Medicine. 1991;324(6):370-6.
2. Wilson RM, Runciman WB, Gibberd RW, Harrison BT, Newby L, Hamilton JD. The Quality in Australian Health Care Study. The Medical Journal of Australia. 1995;163(9):458-71.
3. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospital patients. The Journal of the American Medical Association. 1991;266(20):2847-51.
4. Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. The Journal of the American Medical Association. 1995;274(1):29-34.
5. Cooper JB, Newbower RS, Long CD, McPeek B. Preventable anesthesia mishaps: a study of human factors. Anesthesiology. 1978;49(6):399-406.
6. Ioannidis JP, Lau J. Evidence on interventions to reduce medical errors: an overview and recommendations for future research. Journal of General Internal Medicine. 2001;16(5):325-34.
7. Leape LL, Lawthers AG, Brennan TA, Johnson WG. Preventing medical injury. Quality Review Bulletin. 1993;19(5):144-9.
8. Weingart SN, Wilson RM, Gibberd RW, Harrison B. Epidemiology of medical error. The British Medical Journal. 2000;320(7237):774-7.
9. Starmer AJ, Sectish TC, Simon DW, Keohane C, McSweeney ME, Chung EY, et al. Rates of medical errors and preventable adverse events among hospitalized children following implementation of a resident handoff bundle. The Journal of the American Medical Association. 2013;310(21):2262-70.
10. Institute of Medicine Committee on Quality of Health Care in A. In: Kohn LT, Corrigan JM, Donaldson MS, editors. To Err is Human: Building a Safer Health System. Washington (DC): National Academies Press (US). Copyright 2000 by the National Academy of Sciences. All rights reserved.; 2000.
11. Andrews LB, Stocking C, Krizek T, Gottlieb L, Krizek C, Vargish T, et al. An alternative strategy for studying adverse events in medical care. Lancet. 1997;349(9048):309-13.
12. Gaba DM. Anaesthesiology as a model for patient safety in health care. The British Medical Journal. 2000;320(7237):785-8.
13. Beecher HK, Todd DP. A study of the deaths associated with anesthesia and surgery: based on a study of 599, 548 anesthesias in ten institutions 1948-1952, inclusive. Annals of Surgery. 1954;140(1):2-35.
14. Haller G, Laroche T, Clergue F. Morbidity in anaesthesia: today and tomorrow. Best Practice & Research Clinical Anaesthesiology. 2011;25(2):123-32.
15. Lundgren AC. Peri-Operative death in two major academic hospitals in Johannesburg, South Africa. [Unpublished Thesis].
16. Catchpole K, Bell MD, Johnson S. Safety in anaesthesia: a study of 12,606 reported incidents from the UK National Reporting and Learning System. Anaesthesia. 2008;63(4):340-6.
17. Gupta S, Naithani U, Brajesh SK, Pathania VS, Gupta A. Critical incident reporting in anaesthesia: a prospective internal audit. Indian Journal of Anaesthesia. 2009;53(4):425-33.
47
18. Runciman W, Webb R, Lee R, Holland R. System Failure: An analysis of 2000 incident reports. Anaesthesia and Intensive Care. 1993;21(5):684-95.
19. Choy YC. Critical incident monitoring in anaesthesia. The Medical Journal of Malaysia. 2006;61(5):577-85.
20. Elder NC, Graham D, Brandt E, Hickner J. Barriers and motivators for making error reports from family medicine offices: a report from the American Academy of Family Physicians National Research Network (AAFP NRN). Journal of the American Board of Family Medicine : JABFM. 2007;20(2):115-23.
21. Barach P, Small SD. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting systems. The British Medical Journal. 2000;320(7237):759-63.
22. Taylor JA, Brownstein D, Christakis DA, Blackburn S, Strandjord TP, Klein EJ, et al. Use of incident reports by physicians and nurses to document medical errors in pediatric patients. Pediatrics. 2004;114(3):729-35.
23. Heard GC, Sanderson PM, Thomas RD. Barriers to adverse event and error reporting in anesthesia. Anesthesia and Analgesia. 2012;114(3):604-14.
24. Kaldjian LC, Jones EW, Rosenthal GE. Facilitating and impeding factors for physicians' error disclosure: a structured literature review. Joint Commission Journal on Quality and Patient Safety / Joint Commission Resources. 2006;32(4):188-98.
25. MacCourt D, Bernstein J. Medical error reduction and tort reform through private, contractually-based quality medicine societies. Am J Law Med. 2009;35(4):505-61.
26. Holden RJ, Karsh BT. A review of medical error reporting system design considerations and a proposed cross-level systems research framework. Human Factors. 2007;49(2):257-76.
27. Kaldjian LC, Jones EW, Rosenthal GE, Tripp-Reimer T, Hillis SL. An empirically derived taxonomy of factors affecting physicians' willingness to disclose medical errors. Journal of General Internal Medicine. 2006;21(9):942-8.
28. Radhakrishna S. Culture of blame in the National Health Service; consequences and solutions. British Journal of Anaesthesia. 2015;115(5):653-5.
29. Vincent CA. Analysis of clinical incidents: a window on the system not a search for root causes. Quality & Safety in Health Care. 2004;13(4):242-3.
30. Staender S. Incident reporting in anaesthesiology. Best practice & research Clinical Anaesthesiology. 2011;25(2):207-14.
31. Leape LL. Reporting of adverse events. The New England Journal of Medicine. 2002;347(20):1633-8.
32. Smith ML, Forster HP. Morally managing medical mistakes. Cambridge quarterly of healthcare ethics : CQ : The International Journal of Healthcare Ethics Committees. 2000;9(1):38-53.
33. Malherbe J. Counting the cost: the consequences of increased medical malpractice litigation in South Africa. South African Medical Journal. 2013;103(2):83-4.
34. Liang BA. Risks of reporting sentinel events. Health Aff (Millwood). 2000;19(5):112-20.
35. Runciman WB. Qualitative versus quantitative research--balancing cost, yield and feasibility. 1993. Quality & Safety in Health Care. 2002;11(2):146-7.
48
36. Pham JC, Girard T, Pronovost PJ. What to do with healthcare incident reporting systems. Journal of Public Health Research. 2013;2(3):e27.
37. Wu AW. Medical error: the second victim. The doctor who makes the mistake needs help too. The British Medical Journal. 2000;320(7237):726-7.
38. Tamuz M, Thomas EJ, Franchois KE. Defining and classifying medical error: lessons for patient safety reporting systems. Quality & Safety in Health Care. 2004;13(1):13-20.
39. Chang A, Schyve PM, Croteau RJ, O'Leary DS, Loeb JM. The JCAHO patient safety event taxonomy: a standardized terminology and classification schema for near misses and adverse events. International Journal for Quality in Health Care : Journal of the International Society for Quality in Health Care / ISQua. 2005;17(2):95-105.
40. Walshe K. Adverse events in health care: issues in measurement. Quality in Health Care : QHC. 2000;9(1):47-52.
49
SECTION 4 – Appendices
Appendix 1 – Postgraduate Approval
50
Appendix 2 – Ethics Approval
51
Appendix 3 – Turnitin Report
52
SECTION 5 - Proposal
Factors influencing adverse event and error reporting in Anaesthesiology
Steven Nel
0204336Y
Supervisor: Juan Scribante
Department of Anaesthesiology
Co-supervisor: Helen Perrie
Department of Anaesthesiology
Co-supervisor: Professor Christina Lundgren
Department of Anaesthesiology
53
5.1 Introduction
Adverse events and errors are still a widespread cause of morbidity and mortality in the
health care environment (1-5), with ample literature detailing that many of these errors are
often preventable (5-9).
Adverse events and errors have been increasingly recognised and focused upon since the
report released in 1999 by the Institute of Medicine titled “To Err is Human: Building a safer
health system”. The report highlighted that there was an “epidemic of medical errors”
occurring in health care, and advocated for the provision of safer care for patients. The
report showed that between 44,000 to 98,000 patients die in hospitals in the United States
yearly due to medical errors that were potentially preventable (10).
Rates of adverse events or errors vary in the literature, but range from approximately 16%
(2) to as high as nearly 46% of patients admitted (11). Serious adverse events were found to
be as high as 13 to 17% of admissions (2, 11).
A current trend within the medical community is to maximise patient safety and reduce
medical errors or harm to patients (10). The field of anaesthesia is known to be a leader in
good safety practices, having initially pioneered adverse event reporting in the medical
domain using lessons learnt from the aviation industry (12), as well as having shown a
considerable decline in mortality rates over the last 50 years (13-17).
In an effort to reduce adverse events, there has been research into improving patient safety
by exploring any factors that may potentially lead to harm. Evidence supports the fact that a
large percentage of adverse events are from “system failures” and “organisational factors”,
which are potentially avoidable. (18)
As a result of the high rate of preventable errors, many medical institutions and societies
have instituted formal adverse event and error reporting systems, as a way to improve
patient safety by allowing for learning from “near misses” and “adverse events” (19).
Benefits of such systems are that analyses can be done on various incidents or adverse
events, which can be used for further learning, and to facilitate the framework for robust
prevention strategies, possibly leading to reduced adverse events in the future. It also
creates a societal culture of reporting, which is currently lacking within the medical field
when compared to other environments or industries where errors can have serious
consequences. (16, 20, 21)
54
Even though many institutions have adverse event reporting systems in place, there is still a
significant problem with under-reporting of adverse events and errors (16, 22). Reasons for
this have been noted to include factors such as time available to report, lack of anonymity,
lack of clarity of definitions of adverse events, and perceived seriousness of the event itself.
Some observational studies have shown that up to 90% of adverse events and errors are
not being reported. (20, 23)
Heard et al (23) in 2012 researched the barriers to adverse event and error reporting within
the Australian and New Zealand context, as well as describing factors which would improve
reporting. The authors found that a significant barrier to reporting was fear of blame by
colleagues. They also described further barriers to reporting when doctors had made an
error that led to an adverse event as compared to doctors who had not made an error even
though an adverse event had occurred.
There is paucity of information available regarding adverse event and error reporting
systems implemented currently in South Africa. The Nelson R Mandela School of Medicine
and the University of KwaZulu-Natal have published guidelines regarding the reporting of
adverse events and errors (24), and The Council for Health Service Accreditation of
Southern Africa have instituted a pilot project to run in 45 public institutions in the Free State
which started in 2007. This pilot project is modelled on Australian Incident Management
System that was initiated in 1996 (25). These pilot projects are however on a small scale,
and there is no information regarding their success
5.2 Problem statement
It has been shown that adverse event and error-reporting systems are an integral part of a
health care safety culture, and have the potential to reduce harm to patients, yet formal
adverse event and error reporting systems are not widely available in South Africa. There is
also no evidence that the barriers to reporting of adverse events and errors as described in
the literature would be applicable to developing countries such as South Africa, who face
their own unique health system challenges, nor whether they would be applicable to the
Department of Anaesthesiology at Wits.
5.3 Aim
The aim of this study is to describe anaesthetists perceived barriers to the reporting of adverse
events and errors, and factors that may promote or encourage this reporting in the Department
of Anaesthesiology at Wits.
55
5.4 Objectives
The objectives of this study will be to:
• describe knowledge of existing adverse event or error reporting systems that are
utilised in daily practice
• describe the perceived barriers to the reporting of adverse events and errors
• describe the perceived barriers to the reporting of an adverse event in a theoretical
clinical scenario where an error was made by the anaesthetist, compared to a
theoretical clinical scenario where an adverse event was present but no error made
by the anaesthetist
• describe the factors that may promote or encourage reporting of an adverse event or
error.
5.5 Research assumptions
The following definitions will be used in the study
Anaesthetist: is any qualified doctor working in the Department of Anaesthesiology
including medical officers, registrars and consultants.
Medical officer: is a qualified doctor practising in the Department of Anaesthesiology under
specialist supervision. Medical officers with more than 10 years of experience are career
medical officers and are regarded as consultants.
Registrar: is a qualified doctor that is registered with the Health Professions Council of
South Africa as a trainee anaesthetist.
Consultant: is an anaesthesiologist who has completed all criteria and passed the required
South African College of Medicine examinations, or equivalent. They are regarded as
specialists in the field. Career medical officers are included in this definition.
Error: “the failure of a planned action to be completed as intended (i.e. error of execution) or
the use of a wrong plan to achieve an aim (i.e. error of planning). Errors may be errors of
commission or omission, and usually reflect deficiencies in the systems of care”. (26)
Adverse event: “an injury related to medical management, in contrast to complications of
disease. Medical management includes all aspects of care, including diagnosis and
treatment, failure to diagnose or treat, and the systems and equipment used to deliver care.
56
Adverse events may be preventable or non-preventable”. (26)
Preventable adverse event: “an adverse event caused by an error or other type of systems
or equipment failure” (26).
Near-miss or close call: “serious error or mishap that has the potential to cause an adverse
event but fails to do so because of chance or because it is intercepted. Also called potential
adverse event.” (26)
Adverse drug event: “a medication-related adverse event” (26).
5.6 Demarcation of study field
The study will be conducted in the Department of Anaesthesiology, affiliated to the Faculty of
Health Sciences of the University of the Witwatersrand. The staff complement of the
department is 22 medical officers, 112 registrars and 74 consultants. The following hospitals
are affiliated to the university.
• Charlotte Maxeke Johannesburg Academic Hospital, a 1200 bed central hospital.
• Chris Hani Baragwanath Academic Hospital, a 2888 bed central hospital.
• Helen Joseph Hospital, a 500 bed tertiary hospital.
• Rahima Moosa Mother and Child Hospital, a 338 bed regional hospital.
5.7 Ethical considerations
Approval to conduct the study will be obtained from the Human Research Ethics Committee
(Medical) and the Post Graduate Committee of the University of the Witwatersrand.
This study uses an anonymous self-administered questionnaire (Appendix 1). The
questionnaire will be voluntary and consent is implied on completion of the questionnaire.
The researcher will approach participants at departmental academic meetings, explain the
study and invite them to take part. If they agree the researcher will give them a participant’s
information letter (Appendix 2) with the questionnaire.
Data will be collected without identifying information, and will be assigned a study number,
thus maintaining anonymity. Completed questionnaires will be placed in sealed boxes.
Confidentiality will be ensured, as only the researcher and supervisors will have access to
the raw data.
57
Data will be stored securely for six years after completion of the study. Study will be
conducted according to the principles of the Declaration of Helsinki (27) and the South
African Guidelines for Good Clinical Practice (28).
5.8 Methodology
5.8.1 Research design
Burns and Grove (29) describe a research design as the framework for a study. According to
Brink (30), a research design determines the methods by which the researcher obtains
subjects, collects data and interprets results.
A prospective, contextual, descriptive research design will be followed in this study.
A prospective study is one that measures variables that occur during the course of the study
(30). This is a prospective study, as data will be collected at the time the study takes place.
A contextual study is one which refers to a specific group or population, defined by De Vos
et al (31) as a “small-scale world”. The “small-scale world” can be for example a ward,
intensive care unit or a clinic. This study is contextual because research will be done with
anaesthetists working at hospitals affiliated to the University of the Witwatersrand.
According to Brink (30), a descriptive study is one in which a population’s characteristics are
being described, so as to answer a specific question about the population, without
attempting to establish a causal link. The factors influencing adverse event and error
reporting in anaesthesia will be described.
5.8.2 Study population
The study population consists of all anaesthetists working in the Department of
Anaesthesiology.
5.8.3 Study sample
Sample method
In this study a convenience sampling method will be used, as is suitable for a descriptive
design (29). Convenience sampling involves the sampling of respondents who are readily
available to the researcher (30). This study will sample participants who attend departmental
academic meetings in the Department of Anaesthesiology.
58
Sample size
The sample size will be realised by the response rate. At the time of the study will take place
there should be 208 eligible anaesthetists for the study. Questionnaires will be administrated
to the entire accessible population who are present at academic meetings. At the time of
data collection, it is estimated that 42 (20%) anaesthetists will be inaccessible due to leave,
out of town rotations, etc. A response rate of 60% (100 questionnaires) is considered as
acceptable, but a response rate of 80% (132 questionnaires) will be targeted.
Inclusion and exclusion criteria
The inclusion criterion for this study is all anaesthetists working in the Department of
Anaesthesiology who are willing to partake in the study.
The exclusion criteria for this study are:
• interns
• anaesthetists on annual or sick leave.
5.9 Collection of data
5.9.1 Development of questionnaire
Based on an extensive literature review a questionnaire was identified from a study
published in 2012 by Heard et al (23) in Australia. The questionnaire was modified to
contextualise it to the South African environment. The questionnaire was then reviewed by
three senior anaesthesiologists to achieve face and content validity. Following consultation,
minor corrections were made.
The self-administered questionnaire (23) (Appendix 1) consists of five sections. Questions in
section 2, 3 and 4 are statements listed with a 5-point Likert scale ranging from “strongly
agree” to “strongly disagree”.
Section 1 will focus on the demographics characteristics of the respondents and will include:
• gender
• age group
• professional designation
• years of anaesthetic experience
59
• knowledge of existing adverse event or error reporting systems that are utilised in
participants’ daily practice.
Section 2 will specifically explore attitudinal and emotional factors that could influence
whether an unspecified adverse event caused by an error would be reported. It contains 13
statements; 10 statements centred around the theme of attitudinal barriers and 3 statements
based on fear and anxiety barriers.
Section 3 will focus on reporting an adverse event of anaphylaxis, with or without an error as
its cause. It will contain two case scenarios with two different endings. The first case
scenario will describe an anaphylaxis due to antibiotic administration for a patient who gave
no history of any drug allergy. The second case scenario will be similar to the first case
scenario, but will state that the anaesthetist will realise afterwards that he/she have given a
particular antibiotic in error, after noting that the patient had an allergy to that particular
antibiotic. There will be 17 statements for each scenario to rate as barriers to reporting the
adverse event of anaphylaxis.
Section 4 will focus on strategies that may improve the reporting of adverse events and
errors. There will be 17 statements on factors that may promote the reporting of adverse
events and errors amongst anaesthetists.
Section 5 will be for participant’s free text, allowing for general comments regarding the
questionnaire, as well as any further factors that may influence the reporting of adverse
events and errors.
5.9.2 Data collection
The researcher will approach the convenor at the department academic meetings, and ask
to address the meeting. Information will be provided to the meeting attendees regarding the
research, and the researcher will invite them to take part in the study. Those who are
interested will receive an information letter (Appendix 2) and the questionnaire (Appendix 1).
The researcher will be available to answer any questions, and the completed questionnaires
will be returned in a sealed box.
Blank questionnaires that are returned will be assigned a number and used for response
rate calculation but not for data analysis.
60
5.9.3 Data analysis
Data will be captured onto spreadsheets using Microsoft Excel 2011.
The statistical program Graphpad Prism v5.02 will be used to analyse data, in consultation
with a biostatistician. Descriptive statistics will be used. Categorical variables will be
summarised using frequencies and percentages. The Likert scale will be treated as ordinal
data and Wilcoxon matched pairs will be used to compare variables. A P value of <0.05 will
be considered to be statistically significant.
5.10 Significance of the study
There is considerable evidence within the literature demonstrating that the introduction of an
adverse event and error reporting system is associated with the potential for reduced harm,
a view that is supported by the World Health Organisation (26) and the Institute of Medicine
(10) in the United States.
There is also evidence from studies by Heard et al (23) and other authors (32-37) that define
some of the barriers to adverse event and error reporting. However, there is wide variability
between the various authors’ findings. It is also important to note that the research from
these authors has been done in developed countries. South Africa as a developing country
potentially has different health care challenges, and thus research from Heard et al (23) and
other authors may not translate to the effective implementation of an adverse event and
error reporting system in South Africa.
This study seeks to describe the barriers to adverse event and error reporting that are
applicable in the Department of Anaesthesiology at Wits. This study also seeks to describe
which factors would facilitate a greater likelihood of reporting adverse events. The results of
this study may direct further research aimed at creating a framework for the implementation
of a formal adverse event and error reporting system in South Africa, or allow for the
strengthening of adverse event and error reporting systems that may already be present.
5.11 Validity and reliability of the study
Validity is defined as the accuracy of an instrument to perform an intended measurement.
Reliability refers to the consistency with which an instrument yields the same results (30).
The validity and reliability of this study will be maintained by:
61
• using a validated questionnaire with content and face validity
• placing completed questionnaires into a sealed box, thus facilitating a non-
threatening environment and anonymity
• checking 10% of the data entered to ensure quality of data entry
• consulting with a biostatistician for data analysis.
5.12 Potential limitations of the study
Limitations are defined as restrictions or problems that may decrease the application of the
findings to the general population (29).
The potential limitations of this study are that:
• the study is contextual to the Department of Anaesthesiology at Wits, and thus may
not be able to be generalised to other anaesthesiology departments
• a convenience sample will be used in this study. This may not adequately represent
the perceptions of all anaesthetists in the department.
62
5.13 Project outline
Activity Oct
2014
Nov
2014
Jan
2015
Feb
2015
Apr
2015
May
2015
June
2015
Jul
2015
Aug
2015
Sept
2015
Oct
2015
Proposal
preparation
Chapter 1, 2, 3
Proposal
submission
Ethics
approval
Postgraduate
approval
Data collection
Data analysis
Chapter 4, 5
Submission
5.14 Financial plan
Budget
Item Number Cost Total
Printing 1200 R1/page R1 200
Binding 3 R150 R 450
Total R1 650
The Department of Anaesthesiology will bear the cost of printing and paper for the proposal,
ethics and postgraduate approvals.
63
5.15 References
1. Brennan TA, Leape LL, Laird NM, Hebert L, Localio AR, Lawthers AG, et al. Incidence of adverse events and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. The New England journal of medicine. 1991;324(6):370-6.
2. Wilson RM, Runciman WB, Gibberd RW, Harrison BT, Newby L, Hamilton JD. The Quality in Australian Health Care Study. The Medical journal of Australia. 1995;163(9):458-71.
3. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospital patients. The Journal of the American Medical Association. 1991;266(20):2847-51.
4. Bates DW, Cullen DJ, Laird N, Petersen LA, Small SD, Servi D, et al. Incidence of adverse drug events and potential adverse drug events. Implications for prevention. ADE Prevention Study Group. The Journal of the American Medical Association. 1995;274(1):29-34.
5. Cooper JB, Newbower RS, Long CD, McPeek B. Preventable anesthesia mishaps: a study of human factors. Anesthesiology. 1978;49(6):399-406.
6. Ioannidis JP, Lau J. Evidence on interventions to reduce medical errors: an overview and recommendations for future research. Journal of general internal medicine. 2001;16(5):325-34.
7. Leape LL, Lawthers AG, Brennan TA, Johnson WG. Preventing medical injury. Quality Review Bulletin. 1993;19(5):144-9.
8. Weingart SN, Wilson RM, Gibberd RW, Harrison B. Epidemiology of medical error. The British Medical Journal. 2000;320(7237):774-7.
9. Starmer AJ, Sectish TC, Simon DW, Keohane C, McSweeney ME, Chung EY, et al. Rates of medical errors and preventable adverse events among hospitalized children following implementation of a resident handoff bundle. The Journal of the American Medical Association. 2013;310(21):2262-70.
10. Institute of Medicine Committee on Quality of Health Care in A. In: Kohn LT, Corrigan JM, Donaldson MS, editors. To Err is Human: Building a Safer Health System. Washington (DC): National Academies Press (US)
Copyright 2000 by the National Academy of Sciences. All rights reserved.; 2000.
11. Andrews LB, Stocking C, Krizek T, Gottlieb L, Krizek C, Vargish T, et al. An alternative strategy for studying adverse events in medical care. Lancet. 1997;349(9048):309-13.
12. Gaba DM. Anaesthesiology as a model for patient safety in health care. The British Medical Journal. 2000;320(7237):785-8.
13. Beecher HK, Todd DP. A study of the deaths associated with anesthesia and surgery: based on a study of 599, 548 anesthesias in ten institutions 1948-1952, inclusive. Annals of surgery. 1954;140(1):2-35.
14. Haller G, Laroche T, Clergue F. Morbidity in anaesthesia: today and tomorrow. Best practice & research Clinical anaesthesiology. 2011;25(2):123-32.
15. Lundgren AC. Peri-Operative death in two major academic hospitals in Johannesburg, South Africa. [Unpublished Thesis]. In press 2011.
64
16. Catchpole K, Bell MD, Johnson S. Safety in anaesthesia: a study of 12,606 reported incidents from the UK National Reporting and Learning System. Anaesthesia. 2008;63(4):340-6.
17. Gupta S, Naithani U, Brajesh SK, Pathania VS, Gupta A. Critical incident reporting in anaesthesia: a prospective internal audit. Indian journal of anaesthesia. 2009;53(4):425-33.
18. Runciman W, Webb R, Lee R, Holland R. System Failure: An analysis of 2000 incident reports. Anaesthesia and intensive care. 1993;21(5):684-95.
19. Choy YC. Critical incident monitoring in anaesthesia. The Medical journal of Malaysia. 2006;61(5):577-85.
20. Elder NC, Graham D, Brandt E, Hickner J. Barriers and motivators for making error reports from family medicine offices: a report from the American Academy of Family Physicians National Research Network (AAFP NRN). Journal of the American Board of Family Medicine : JABFM. 2007;20(2):115-23.
21. Barach P, Small SD. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting systems. The British Medical Journal. 2000;320(7237):759-63.
22. Taylor JA, Brownstein D, Christakis DA, Blackburn S, Strandjord TP, Klein EJ, et al. Use of incident reports by physicians and nurses to document medical errors in pediatric patients. Pediatrics. 2004;114(3):729-35.
23. Heard GC, Sanderson PM, Thomas RD. Barriers to adverse event and error reporting in anesthesia. Anesthesia and analgesia. 2012;114(3):604-14.
24. Mahomed O. Adverse events monitoring and reporting guidelines [26 March 2016]. Available from: http://www.kznhealth.gov.za/family/Adverse-events-monitoring-reporting-guidelines.pdf.
25. Improving the safety of patient care: The Council for Health Service Accreditation of Southern Africa; [cited 2016 26 March]. Available from: http://www.cohsasa.co.za/improving-safety-patient-care.
26. WHO Draft Guidelines for Adverse Event Reporting and Learning Systems. Geneva, Switzerland: World Health Organization, 2005.
27. World Medical Association. WMA Declaration of Helsinki-Ethical principles for medical research involving human subjects. Brazil: World Medical Association, 2013.
28. Deprtment of Health. Guidelines for Good Practice in the Conduct of Clinical Trials with Human Participants in South Africa. Pretoria, South Africa: 2006.
29. Burns N, Grove S. The Practice of Nursing Research. Missouri: Saunders Elselvier; 2009.
30. Brink H, van der Walt C, van Rensburg G. Fundementals of Research Methodology for Healthcare Professionals. Cape Town: Juta & Company Ltd.; 2012.
31. De Vos A, Strydom H, Fouche C, Poggenpoel M, Schurink E, Schurink W. Research at Grass Roots. Pretoria: Van Schaik; 1998.
32. Holden RJ, Karsh BT. A review of medical error reporting system design considerations and a proposed cross-level systems research framework. Human factors. 2007;49(2):257-76.
33. Suresh G, Horbar JD, Plsek P, Gray J, Edwards WH, Shiono PH, et al. Voluntary anonymous reporting of medical errors for neonatal intensive care. Pediatrics. 2004;113(6):1609-18.
65
34. Jeffe DB, Dunagan WC, Garbutt J, Burroughs TE, Gallagher TH, Hill PR, et al. Using focus groups to understand physicians' and nurses' perspectives on error reporting in hospitals. Joint Commission journal on quality and safety. 2004;30(9):471-9.
35. Leape LL. Reporting of adverse events. The New England journal of medicine. 2002;347(20):1633-8.
36. Wu AW. Medical error: the second victim. The doctor who makes the mistake needs help too. The British Medical Journal. 2000;320(7237):726-7.
37. Kaldjian LC, Jones EW, Rosenthal GE. Facilitating and impeding factors for physicians' error disclosure: a structured literature review. Joint Commission journal on quality and patient safety / Joint Commission Resources. 2006;32(4):188-98.
66
Appendix 1: Questionnaire (23)
67
68
69
70
71
72
Appendix 2: Information sheet
Dear Colleague,
Hello, my name is Steven Nel, and I am an anaesthesiology registrar in the Wits Department of
Anaesthesiology. I would like to invite you to participate in my MMed research study titled “Factors
influencing adverse event and error reporting in Anaesthesiology”. This study was approved by the
Human Research Committee (Medical) (M150102)
The study aims to describe the perceived barriers to the reporting of adverse events and errors amongst
South African anaesthetists, as well as to determine which factors might facilitate enhanced reporting
practices amongst anaesthetists.
Participation is on a voluntary basis and consent is implied on completion of a questionnaire. All
information given will remain strictly confidential and anonymous, as no personal details will need to be
provided to compete the questionnaire. There will be no penalty for not participating in this study.
All questionnaires, whether completed or not, should please be given back to the researcher. All
questionnaires will be given a unique number once back with the researcher, that can in no way identify
the participant. Only my supervisors and I will view the completed surveys, thereby ensuring
confidentiality. The questionnaire should not take longer than 10 minutes to complete.
No incentives will be provided for the completion of the questionnaire. Completion of this questionnaire
will however assist in identifying barriers to adverse event and error reporting, and could help implement
a system of reporting in the future, to the benefit of all anaesthetists.
Before completion of the survey, please ensure that you understand the above information.
Your time is greatly appreciated. Any questions regarding this study can be directed to the following
people:
• ProfessorCleaton-Jones(chairpersonoftheHREC):(011)717-1234
• StevenNel(researcher):0842398007
Sincerely,
Steven Nel
